Bernstein isa truepioneer indeveloping practical approaches tocontrolling
a devastating disease that isgrowing at epidemic proportions in thiscountry."
—Barry Sears, PhD, author of The Zone
r r?"
Richard K. Bernstein, MD
Dr. Bernstein's
Diabetes Solution
Theories, no matter how pertinent,
Cannot eradicate the existence of facts.
—Jean Martin Charcot
Dedicated to the Memory of My Dear Friends
Heinz I. Lippmann, MD,
Samuel M. Rosen, MD,
Who fervently believedthat people with diabetes are
entitled to the sameblood sugars asnondiabetics
Dr. Bernstein's
Diabetes Solution
Newly Revised and Updated
The Complete Guide
to Achieving Normal Blood Sugars
Richard K. Bernstein, MD,
Foreword by FrankVinicor, MD, MPH
Recipesby Karen A. Weinstockand Timothy J. Aubert, CWC
i 837
NewYork Boston London
Copyright©1997,2003,2007 by Richard K. Bernstein, MD
Glucograf* is a registered trademarkof Richard K. Bernstein, MD
All rightsreserved. Exceptas permittedunderthe U.S.CopyrightAct of 1976,
no partof this publication maybe reproduced, distributed, or transmitted in
anyformorby anymeans,or storedinadatabase orretrieval system,without
the prior written permissionof the publisher.
Little, Brown and Company
Hachette Book Group
237 Park Avenue, NewYork, NY 10017
Visit our Web site at www.HachetteBookGroup.com
Originally published in hardcover by Little, Brown andCompany, 1997
Newlyrevisedand updatededition, March 2007
Little, Brownand Company is a divisionof Hachette BookGroup, Inc.
The Little, Brown name and logo aretrademarksof Hachette Book Group, Inc.
Illustrations by Terry Eppridge
Author's Note
This book is not intended as a substitute for professional medical care.
The reader should regularly consult a physician for all health-related
problems and routine care.
The author is grateful for permission to include the following previously
Figure 1-3. Reproduced from the Journal of Clinical Investigation, 1967;
46:1549-1557. By permissionof The AmericanSociety for Clinical Investi
Figure 9-1. Reproduced fromJournal of theAmerican Dietetic Association,
1995; 45:417-420. Copyright © by The American Dietetic Association.
Reprintedby permissionof The AmericanDietetic Association.
Figure 19-1. Reproduced fromHumalog PI. Reprinted by permission of Eli
Lilly and Company.
Library ofCongress Cataloging-in-Publication Data
Bernstein, Richard K.
Dr. Bernstein's diabetes solution: the complete guide to achieving normal
blood sugars / Richard K. Bernstein; foreword by FrankVinicor; recipesby
Karen A. Weinstock and Timothy J. Aubert. — Newly rev. and updated,
p. cm.
Doctor Bernstein's diabetes solution
Diabetes solution
Includes index.
ISBN 978-0-316-16716-1
1. Diabetes— Popularworks. 2. Blood sugar monitoring — Popular
works. I. Title. II. Title: Doctor Bernstein's diabetes solution. III. Tide:
Diabetes solution.
RC660.4B464 2007
616.4'62 —dc22 2006026483
10 9 8 7 6 5
Printed in the United States of America
Forewordby FrankVinicor,MD, MPH vii
Prefaceto the NewlyRevised and Updated Edition ix
MyLife with Diabetes: Well Beyond a HalfCenturyand Counting xii
Acknowledgments xxi
Before and After: Fourteen Patients Share Their Experiences 3
Before You Start
1. Diabetes: The Basics 33
2. Tests: Baseline Measures of Your Disease and Risk Profile 52
3. Your Diabetic Tool Kit: Supplies YouWill Need and Where
to Get Them 66
4. Howand When to MeasureBloodSugar 75
5. RecordingBlood Sugar Data: Usingthe Glucograf III Data Sheet 83
6. Strange Biology: Phenomena Peculiar to DiabetesThat
CanAffect BloodSugar 91
7. The Laws of Small Numbers 102
8. Establishinga Treatment Plan: The Basic Treatment Plans
and HowWe Structure Them 109
9. The BasicFood Groups, or Much of What You've Been
Taught About Diet Is Probably Wrong 123
10. Diet Guidelines Essential to the Treatment ofAll Diabetics 138
11. Creating a Customized Meal Plan 167
12. Weight Loss — If You're Overweight 184
13. How to Curb Carbohydrate Cravingor Overeating
Using Self-Hypnosis or Low-Risk Medications 196
14. UsingExercise to Enhance InsulinSensitivity 211
15. Oral Insulin-SensitizingAgents,Insulin-Mimetic Agents,
and Other Options 235
16. Insulin: The Basics of Self-Injection 249
17. Important InformationAboutVarious Insulins 264
18. SimpleInsulin Regimens 276
19. Intensive Insulin Regimens 284
20. Howto Preventand Correct LowBloodSugars 317
21. Howto Cope with Dehydration, DehydratingIllness, and Infection 344
22. DelayedStomach-Emptying: Gastroparesis 357
23. Routine Follow-upVisitsto YourPhysician 381
24. What You Can ExpectfromVirtuallyNormal BloodSugars 385
Your Diabetic Cookbook
25. Recipes for Low-Carbohydrate Meals 391
AppendixA:What About theWidelyAdvocated DietaryRestrictions
on Fat, Protein, and Salt,and the Current High-FiberFad? 443
Appendbc B: Don't Permit Hospitalization or Lengthy Outpatient
Procedures to ImpairYour Blood SugarControl 462
Appendix C: Drugs That MayAffect BloodGlucoseLevels 465
Appendix D: Foot Care for Diabetics 473
Appendix E: Polycystic OvarianSyndrome 477
Glossary 482
RecipeIndex 493
General Index 495
by FrankVinicor, MD, MPH
Director, Division of Diabetes Translation
National Center for Chronic Disease Prevention
and Health Promotion
Centers for Disease Control and Prevention
Atlanta, Georgia
Weare learninga lot about diabetes — especially during the
past five to ten years. This accumulation of new knowledge
is both encouraging and at the sametime verychallenging.
On the "challenging'' side:
• Diabetes seems to be everywhere and steadily increasing in its
presence. Think about it — 1 in 3 babies born in 2000 will de
velop diabetes in their lifetimes. Every day, about 1,400 people
are diagnosed with diabetes in the United States. And now no
country in the worldis free fromdiabetes, and its growth.
• We now do know how to prevent type 2 diabetes, but today for
type 1 diabetes, neither prevention nor a long-lasting cure is
• Once diabetes is present, good carebased on solid science now
can prevent much of the devastation formerly causedby elevated
blood sugars. But there remains a sizable gap between what we
know to do and how well and widely we are doing it. In other
words, the "translation" of diabetes science into daily practice
still has a way to go.
Nonetheless, in spite of these and other important challenges, we
areallbetter preparedto dealwith diabetes in 2007than we were even
a fewyears earlier, let alone decades ago. Remarkableprogress has oc
curred. Forexample, many people at high risk for type 2 diabetes do
not develop it. Modest weight lossand increased physical activity have
been shown to eliminateor at least delay the development of this type
of diabetes by 60-70 percent — regardless of race, ethnicity, or age.
In addition, for both types 1 and 2 diabetes, we now have many
viii Foreword
more effective medications which, when taken appropriately and in
combination with proper nutrition and activity, will result in con
trolled plasma glucose, blood pressure, andblood fats — with definite
reduction in the likelihood of eye, kidney, nerve, and heart problems.
In other words, while the goals of diabetes research still in large part
should be prevention or cure, even now the devastation formerly
caused by this conditiondoes nothave tohappen!
Nowadays, too, we have better ways to follow and keep track of
diabetes — with improved health care systems, better educational
programs, less painful self-monitoring of blood sugars, more quickly
available and accurate glycosylated hemoglobin levels, waysto identify
kidney problemsearly, and so forth. We canknow what is goingon!
So, in fact, we areactually seeingan improvement in diabetes care
in the United States, althoughnot with all people and not yet to an ad
equatelevel or fast enough.
What does all this have to do with Dr. Bernstein and this edition of
Diabetes Solution7. As mentioned earlier, the rate of accumulation of
new diabetes knowledge is quite remarkable and daunting. Yet Dr.
Bernstein stays on top of it all. The care pattern for diabetes has be
come much more complex anddemanding, and Dr. Bernsteinand his
approachhave provedequalto the challenge. In essence, diabetes is in
many ways"lesseasy" than in the past— forthe patient or for his/her
health care professional. There are lots of nutritional approaches to
consider, lots of medications to be used in varying combination, and
often less time within abusy office practiceto make all these wonder
ful advances real and meaningful for people facing diabetes. This
newly revisededition presents the advances in diabetes thinking and
management with passion, compassion, caring, and conviction. Cer
tainly, forsome people, hisapproaches are not easy! But they do reflect
evolvingmedical science aswell ashis personal experiences in manag
ing his own diabetes. He does not ask anyone to do anything that he
himself would not do, and for this I have respect and admiration. He
is offeringto persons challenged by the presence or risk of diabetes a
way to be in charge of the disease. And he is ensuring that important
advances in diabetes scienceget out there nowto make a difference in
people's lives. Take a look! Think about the ideas and suggestions —
they can further our mutual and ongoing effort to prevent, capture,
and control this disease called diabetes.
Preface to the Newly Revised and
Updated Edition
Since the publication of the revised edition of Dr. Bernstein's
Diabetes Solution in 2003, many new developments have oc
curred in the fieldof diabetes research, and as each significant
one has come along,I have furtherrefinedmy techniques for normal
izing blood sugars. This newly revised and updated edition discusses
new oral medications, new insulins, new dietary supplements, new
hardware (tools for the diabetic), and other new products. It also ex
plores newmethods that I havedeveloped for more elegantly control
ling blood sugars.
Exciting new approaches to weight losswill be found here, includ
ing the use of a new, injectablemedication (an amylin analog) that is
wonderfully effective for alleviating carbohydrate craving and over
This newly revisedand updated edition builds upon the prior two
editions of this book and upon my two earlierbooks about diabetes. It
is designedasatool for patientsto be used under the guidanceof their
physicians or diabetes educators. It covers, in a step-by-step fashion,
virtually everything that must be done to keep blood sugars in the
normal range.
In these pagesI attempt to present nearly everything I know about
blood sugar normalization, how it can be accomplished and main
tained.With this book, and with the help of your physicianor diabetes
educator, I hope that you will learn to take control of your diabetes,
whether it's type 1 (juvenile-onset), as mine is, or the much more
common type 2 (maturity-onset) diabetes. To my knowledge, there is
no other book in print addressed strictly to blood sugar control for
both types of diabetes.
x Preface to theNewlyRevisedand UpdatedEdition
This volume contains much material that may be new to many
physicians treating diabetes. It ismy hopethat doctors andhealthcare
professionals will use it, learn from it, and do their best to help their
patients take control of this potentially deadly but controllable
Although thisbook contains considerable backgroundinformation
on diet and nutrition, it is intended primarily as a comprehensive
how-to guideto blood sugar control, including detailed instructions
on techniques for painless insulininjectionand so on. It must, there
fore, leave out other related issues (suchaspregnancy), some of which
require their own volumes. My officetelephone number is listed sev
eral times in this book, and we arealways happy to hear from readers
who seek our latest recommendation for a blood sugar meter, other
equipment, or new medications.
I urgeyoutovisittheWebsite for thisbook,www.diabetes-book.com.
The site contains some of my recent articles, a history of blood sugar
self-monitoring, links to other sites, testimonials from readers who
havetried the program, anopportunity to share your own experiences
in anongoingchat groupfor diabetics andtheirlovedones, andmore.
The site alsopermits you to forward information by e-mail to anyone
you think could benefit fromthis book.
Recent news releases and advertisements have described "develop
ments" and products that are not mentioned here, and you may be
curious about them. If a medication is not discussed here, then it is
likely I have deliberately omitted it as either useless or potentially
harmful, or it was not available when this volume was written. There
are many drugs, oldandnew,usedin the treatmentof diabetes. Some,
like metformin, or Glucophage, are truly wonderful, but others, such
as the sulfonylureas, are insidious and can destroy your body's re
maininginsulin-producing capability, if it hasany. I haveomitted any
thing I think is either too far into the future to be of near-term
consequence or is simply not going to be effective at getting you on
track. I haveneither the time nor the space to attempt to debunk every
"miracle cure" that comes along, most of which are neither miracu
lous nor cures.
Should you become pregnant while on this program, of all the
medications mentioned in this book, metformin, aspirin, and insulin
are the only ones that have been tested in pregnant women. Never
theless, check out all your medications with your obstetrician and
pharmacist — ideally beforeyou become pregnant.
Preface to the Newly Revised andUpdated Edition xi
Manythousands of diabetics have successfully used this program.
Like them, if you, with your physician's help, seriously follow these
guidelines, youshould be able to avoid the discomfort of inappropri
atebloodsugar swings. Youmayevenbe able to prevent or reverse the
development of manyof thegrave complications long associated with
chronically highbloodsugars.
Finally, much of what I will cover in this bookis in direct opposi
tion to the recommendations of the American Diabetes Association
(ADA) andothernational diabetes associations. Why? Because if I had
followed those guidelines, theywould have killed me long ago. Such
conflicts include the low-carbohydrate diet I recommend; the avoid
ance of oral agents (such as sulfonylureas) that burn out surviving
insulin-producing betacells in type 2 diabetics; my utilization of cer
tain nutrients to lower insulin resistance; my preference for certain
insulins over others, which I avoid; my desire to preserve remaining
beta cells (an alien concept to traditional practice); andmy insistence
that diabetics are entitled to the same, normal bloodsugars that non
diabetics enjoy, rather than theADA's current insistence uponhigher
Most important, unlike the ADAguidelines, ours work.
My Life with Diabetes
I do not know of many diabeticswho developedthe illness around
the time I did, in 1946, who are still alive. I know of none who do
not suffer from active complications. The reality is, had I not
taken charge of my diabetes, it's very unlikelythat I'd be aliveand ac
tivetoday. Manymyths surround diet and diabetes,and much of what
is still considered by the average physician to be sensiblenutritional
advicefor diabeticscan, overthe long run, be fatal.
I know, because conventional "wisdom" about diabetes almost
killed me.
I developed diabetes in 1946 at the ageof twelve, and for more than
two decades I was an "ordinary" diabetic, dutifully followingdoctor's
orders and leading the most normal life I could, giventhe limitations
of my disease.
Over the years, the complications from my diabetes became worse
and worse, and like many diabetics in similar circumstances, I faced a
veryearlydeath. I was still alive, but the qualityof mylifewasn't par
ticularly good. I have what is known as type 1, or insulin-dependent,
diabetes, which usuallybegins in childhood (it's also called juvenile-
onset diabetes). Type 1diabetics must takedailyinsulininjectionsjust
to stay alive.
Backin the 1940s, whichwereverymuch still the "dark ages"of di
abetes treatment, I had to sterilize my needles and glass syringes by
boiling them every day, and sharpen my needles with an abrasive
stone. I used a test tube and an alcohol lamp (flame) to test my urine
for sugar. Many of the tools the diabetic can take for granted today
were scarcely dreamed of back then — there was no such thing as a
rapid, finger-stick blood sugar-measuring device, nor disposable in-
My Lifewith Diabetes xiii
sulinsyringes. Still, eventoday, parents of type 1diabetics have to five
with the same fear my parents lived with — that something could go
disastrously wrongand they couldtry to wakeup their childand dis
cover him comatose, or worse. For any parent of atype 1diabetic, this
hasbeen areal and constant possibility.
Because of my chronically elevated blood sugar levels, and the
inability to control them, my growth was stunted, as it is for many
juvenile-onset diabetics evento this day.
Back then, the medical community hadjust learned about the rela
tionship between high blood cholesterol and vascular (blood vessel
and heart) disease. It was thenwidely believed that the cause of high
blood cholesterol was consumption of large amounts of fat. Since
many diabetics, even children, have high cholesterol levels, physi
cians were beginning to assume that the vascular complications of
diabetes — heart disease, kidney failure, blindness, et cetera — were
caused by the fat that diabetics were eating. As aresult, I was put on a
low-fat, high-carbohydrate diet (45 percent of calories were to be
carbohydrates) before such diets were advocated by the American
Diabetes Association or the American Heart Association. Because car
bohydrate raises blood sugar, I had to compensate with very large
doses of insulin, which I injected with a 10 cc"horse" syringe. These
injections were slowandpainful, andeventually they destroyed all the
fatty tissue undertheskinof my thighs. Inspite of thelow-fat diet, my
blood cholesterol became veryhigh. I developed visible signs of this
state — fatty growths on my eyelids andgray deposits around the iris
of eacheye.
During my twenties andthirties, the primeof life for most people,
manyof my body's systems began to deteriorate. I hadexcruciatingly
painful kidney stones, a stone in a salivary duct, "frozen"shoulders, a
progressive deformity of my feet with impaired sensation, and more. I
wouldpoint theseout to my diabetologist (whowas then president of
the American Diabetes Association), but I was inevitably told,"Don't
worry, it hasnothingto do withyourdiabetes. You're doingfine." But
I wasn't doing fine. I nowknowthat most of these problems are com
monplace among those whose diabetes is poorlycontrolled, but then
I was forcedto accept my condition as"normal."
Bythis time I wasmarried. I hadgone to college and trainedasan
engineer. I had small children, andeventhough I wasnot much more
than akid myself, I feltlike anoldman. I hadlost the hairon the lower
parts of my legs, a sign that I had developed peripheral arterial dis-
xiv My Lifewith Diabetes
ease — acomplicationof diabetes that caneventuallyleadto amputa
tion. During a routine exercise stress test, I was diagnosed with car
diomyopathy, which is a replacement of muscle tissue in the heart
with fibrous (scar) tissue — a common cause of heart failure and
deathamongthosewith type 1diabetes.
Even though I was "doing fine," I suffered a host of other compli
cations. My vision deteriorated: I suffered night blindness, micro
aneurysms (ballooning of thebloodvessels inmyeyes), macular edema
(swelling of the central portion of my retinas), andearly cataracts. Just
lyingin bed caused pain in my thighs, due to a common but rarely
diagnosed and barely pronounceable diabetic complication called
iliotibialband/tensor fascialata syndrome. Puttingon aT-shirt was ag
onizingbecause of my frozen shoulders.
I had begun testing my urine for protein and found substantial
amounts of it, asign, I hadread, of advanced kidney disease. In those
days — themiddleandlate 1960s — thelifeexpectancyof atype 1di
abeticwith proteinuria was five years. Backin engineering school, a
classmatehad told me how his nondiabetic sister had died of kidney
disease. Before her death she had ballooned with retained water, and
after I discovered my own proteinuria, I beganto havenightmares of
blowing up like aballoon.
By 1967 I had these and other diabetic complications and clearly
appeared chronically ill andprematurely aged. I hadthree small chil
dren, the oldest only six years old, andwith goodreason was certainI
wouldn't live to see them grown.
At my father's suggestion, I startedworkingout daily at alocal gym.
He thought that if I were to engage in vigorous exercise, I might feel
better. Perhaps exercise would help my body help itself. I did feel
slightly less depressed about my condition — atleast I felt I was doing
something— but I couldn't buildmuscles or getmuch stronger.
Aftertwo years of pumpingiron,I remained a 115-poundweakling,
no matter how strenuously I workedout. It wasat about this time, in
1969, that my wife, a physician, pointed out to me that I had spent
much of my life goinginto, experiencing, or recovering from hypo
glycemia, whichis astate of excessively lowbloodsugar. It was usually
accompanied by fatigue andheadaches, andwas caused by the unpre
dictable action of the large doses of insulin I was taking to cover my
high-carbohydrate diet. During suchepisodes, I became confused and
unruly and snapped at people. These frequent hypoglycemic episodes
had taken their toll upon my parents, and weretakingtheir toll upon
MyLifewithDiabetes xv
my wife and children. The strain on my family was clearlybecoming
Suddenly, in October of 1969, my life turned around.
I hadbeen the research directorofacompany that made equipment
for hospital laboratories, but recentlyI had taken anew job asan offi
cer of a housewares corporation. I was still receiving trade journals
from my old field, and one day I opened the latest issueof a publica
tion calledLabWorld. I came upon an advertisement for a new device
to help hospital emergency rooms distinguish between unconscious
diabetics andunconscious drunksduringthe night,whenlaboratories
were closed. Knowing that an unconscious person wasa diabetic and
not drunk could easily help hospital personnel save his life. What I
stumbled upon was an ad for a blood sugar meter that would give a
reading in 1minute, usingasingle dropof blood.
Since I'd been experiencing many blood sugars that were too low,
andsince the testsI hadbeenperforming on my urinewere whollyin
adequate (sugar that showsup in the urineis already on its wayout of
the bloodstream), I figured that if I knewwhat my blood sugars were,
perhaps I could catch and correct my hypoglycemic episodes before
they made me disorientedand irrational.
I marveled overthe instrument. It hada4-inch galvanometer with
a jeweledbearing, weighed 3 pounds, and cost $650.1 tried to order
one, but the manufacturerwouldn't sell it to patients, only to doctors
and hospitals.
Fortunately, my wife, asI've said, wasa physician, so I orderedone
in her name. I started to measure my blood sugar about 5 times each
day, andsoonsawthat thelevels were on aroller coaster. Engineers are
accustomed to solvingproblems mathematically, but you haveto have
informationto work with.Youhave to knowthe mechanics of aprob
lem in order to solveit, and now, for the first time, I was gaining in
sight into the mechanics and mathematics of my disease. What I
learned from my frequent testing was that my own blood sugars
swungfromlowsof under 40mg/dl to highsof over400mg/dl about
twicedaily. A normalbloodsugar level isabout 85mg/dl.* Small won
der I wassubject to suchvast mood swings.
*Althoughmost medical journalsand textbooks throughout the worldmeasure
blood glucose in mmol/1 (millimoles per liter), most physicians, laboratories,
and bloodglucose metersin the UnitedStates measure blood glucose in mg/dl
xvi MyLifewithDiabetes
In an effort to level my blood sugars, I beganto adjust my insulin
regimen, andwent from one injection a dayto two. I made some ex
perimental modifications to my diet, cutting down on the carbohy
drates to permit me to take less insulin. The veryhighandlowblood
sugar levels became less frequent, but few were normal.
Threeyears after I started measuring my bloodsugar levels, my di
abetic complications were still progressing, andI was stilla115-pound
weakling. Mysense of gaining insight intothelong-termcomplications
of my diabetes haddiminished, and soI ordered acomputer search of
the scientific literatureto see if exercise could prevent diabetic com
plications. In those days, computer searches were not the simple, al
most instantsearches theyare today. In 1972 youmadeyourrequest to
the local medical library, whichmailedit toWashington, DC,whereit
was processed. It took about twoweeks for my $75 printout to arrive.
There were quite a few entries of interest, andI ordered copies of
the original articles. For the most part these were from esoteric jour
nals anddealt with animal experiments. The informationI hadhoped
to find didn't exist. I didn't find a single article pertainingto the pre
vention of diabetic complications by exercise.
What I did find was that such complications had repeatedly been
prevented, andeven reversed, inanimals. Not through exercise, but by
normalizing bloodsugars! To me, thiswas atotal surprise. All of dia
betes treatment was heavily focused in other directions, such aslow-
fat diets, preventing severe hypoglycemia, and preventing apotentially
fatal extreme highbloodsugar condition called ketoacidosis. Thus it
had not occurredto me that keepingblood sugarlevelsascloseto nor
mal as possible for as muchof the timeas possible would makeadif
Excited by my discovery, I showed these reports to my physician,
who was not impressed. "Animals aren't humans," he said, "and be
sides, it's impossible to normalize human blood sugars." Since I had
beentrained as anengineer, not as aphysician, I knewnothingof such
impossibilities, andsince I was desperate, I hadno choice but to pre
tend I was an animal.
I spentthe next year checking my blood sugars 5-8 timeseach day.
Every few days, I'dmakeasmall, experimental change inmy diet orin-
(milligrams perdeciliter). Blood glucose values inthisbookareasa rulegiven in
mg/dl. Ifyou should need totranslate fromonetotheother, 1mmol/1 = 18 mg/dl.
My Lifewith Diabetes xvii
sulin regimen to see whatthe effect would beon my bloodsugar. If a
change brought animprovement, I'dretain it. If it madebloodsugars
worse, I'd discard it. I discovered that 1 gram of carbohydrate raised
my blood sugar by 5 mg/dl, and Vi unit of the old beef/pork insulin
loweredit by 15mg/dl.
Within a year, I had refined my insulin and diet regimen to the
point that I hadessentially normal bloodsugars around the clock. Af
ter years of chronic fatigue and debilitating complications, almost
overnight I was no longer continually tired or "washed out." People
commented that my gray complexion was gone. After years of sky-
high readings, my serum cholesterol and triglyceride levels had now
not only dropped, but were at thelowendof the normal ranges.
I started to gain weight, and at last I was able to build muscle as
readily as nondiabetics. My insulin requirements dropped to about
one-third of what they had been a year earlier. With the subsequent
development of human insulin, my dosage dropped to less than one-
sixthof the original. The painful, slow-healing lumps the injections of
large doses of insulin left under my skin disappeared. The fatty
growths on my eyelids from high cholesterol vanished. My digestive
problems (chronic burningin my chest andbelching after meals) and
theproteinuria thathadsoworried meeventuallyvanished. Today, my
results fromeventhe most sensitive kidney function tests are all nor
mal. I recently discovered that even the calcified muscle liningthe ar
teries in my legs has normalized. As chief of the peripheral vascular
disease clinic of a major medical school, I had been teaching physi
cians that acurefor this Monckeberg's atherosclerosis wasimpossible.
I proved myself wrong. My deformed feet, the droopyeyelids, andthe
lossof hair on my lowerlegs are not reversible and still remain.
I hadthe newsensation of beingthebossof my ownmetabolic state,
andbegan to feel the same sense of accomplishment andreward I had
in engineering when I solved a difficult problem. I had taught myself
howto make my blood sugars whatever I wantedthem to be and was
no longer on the roller coaster. Things were finally under my control.
Backin 1973,1 felt quiteexhilarated with my success, and I felt that
I was on to something big. Since getting the results of my computer
search, I hadbeenasubscriber to all of the English-language diabetes
journals, andnone of themhadmentioned the need for normalizing
blood sugarsin humans.
In fact, everyfew months I'dread another article saying that blood
sugarnormalization wasn't even remotely possible. How was it that I,
xviii My Lifewith Diabetes
an engineer,had figured out how to do what was impossible for med
icalprofessionals? I wasdeeplygrateful forthe fortuitous combination
of events that had turned my life, my health, and my family around
and put me on the right path.At the veryleast, I felt, I wasobligedto
share my newfoundknowledge with others. Millions of "ordinary" di
abetics wereno doubt sufferingneedlessly, as I had. I was sure that all
physicians treating diabetes wouldbe thrilledto learnhowto so easily
prevent and possibly reverse the grave complications of this disease.
I hoped that if I could tell the world about the techniques I had
stumbled upon, physicians would adopt them for their patients. So I
wrote an article detailing my discoveries. I sent a copy to Charles
Suther, who was then in charge of marketing diabetes products for
Ames Division of Miles Laboratories, the company that made my
blood glucose meter.He gave me the only encouragement I received in
this newventure, and arranged for one of his company's medical writ
ers to edit the article for me.
I submitted it and its revisions to many medical journalsover a pe
riod of years — a period duringwhich I was continually improving in
health, and continually provingto myself and my family, if to no one
else,that my methods werecorrect. The rejectionletters I receivedare
testimony that peopletend to ignore the obviousif it conflictswith the
orthodoxy of their earlytraining. Typical rejectionlettersreadin part:
"Studies are not unanimous in demonstrating a need for 'fine con
trol' " (the NewEngland Journal ofMedicine), or "How many patients
would use the electric device for measurement of glucose, insulin,
urine, etc.?" (Journal of theAmerican Medical Association). As amatter
of fact, since 1980, when these "electric devices" finally were made
available to patients, the worldwide market for blood glucose self-
monitoring supplies has come to exceed $4 billionannually. Look at
the arrayof blood glucose meters in any pharmacyand you can get an
ideaof just how many patientsuse, and will use, the "electricdevice."
Tryingto coverseveral routessimultaneously, I joinedthe major lay
diabetes organizations, in the hope of moving up through the ranks,
where I could get to know physicians and researchers specializing in
the disease. This met with mediocre success. I attended conventions,
worked on committees, andbecameacquaintedwith many prominent
diabetologists. In this country, I met only three physicianswho were
willingto offer their patients the opportunity to put these new meth
ods to the test.
My Lifewith Diabetes xix
Meanwhile, Charlie Suther was traveling around the country to
university research centers with copies of my unpublished article,
whichby now hadbeen typeset and privately printedat my expense.
The rejection by physicians specializing in diabetes of the concept of
blood sugar self-monitoring, even though essential to blood sugar
control,wassointense,however, that the managementofhis company
had to turn down the idea of makingmeters available to patients un
til many years later. His company andothers couldclearly have prof
ited fromthe sale ofbloodglucose metersandtest strips. However, the
backlash fromthe medical establishment prevented it on anumber of
counts. It was unthinkable that patientsbe allowedto "doctor" them
selves. They knew nothing of medicine— and if they could, how
would doctors earn a living? In those days, patients visited their doc
tors once a month to "get a blood sugar." If they could do it at home
for 25 cents (in those days), why paya physician? But almost no one
believed therewas anyvalue to normalblood sugars anyway. In some
respects, blood glucose self-monitoring still remains a serious threat
to the incomes of many physicians who specialize in the treatment of
the symptoms of diabetes and not the disease. Drop into your neigh
borhoodophthalmologist's office andyouwill find the waiting room
three-quarters filled with diabetics, many of whom arewaiting forex
pensive fluorescein angiographyor laser treatment.
With Suther's backingin the formof free supplies, by 1977 I was
able to get the first of two university-sponsoredstudies started in the
NewYorkCity area. Theseboth succeeded in reversing early compli
cations in diabetic patients. As a result of our successes, the two uni
versities separately sponsored the world's first two symposia on blood
glucose self-monitoring. By this time I wasbeing invited to speak at
international diabetes conferences, but rarely at meetingsin theUnited
States. Curiously, more physicians outside the United States seemed
interested in controlling blood sugar than did their American col
leagues. Some ofthe earliest converts to bloodglucose self-monitoring
were from Israel and England.
By 1978, perhaps as a result of Charlie Suther's efforts, a few addi
tionalAmericaninvestigators were tryingour regimen or variations of
it. Finally, in 1980, manufacturers began to release blood glucose me
ters for use by patients.
This "progress" was entirely too slow for my liking. I knew that
while the medical establishment was dallying there were diabetics dy-
xx My Lifewith Diabetes
ingwhoselives couldhave beensaved. I knewalso that thereweremil
lions of diabetics whose quality of life could be vastly improved. Soin
19771 decidedto giveup my jobandbecome aphysician — I couldn't
beat 'em, so I had to join 'em. This way, with an MD after my name,
my writings might be published, and I could pass on what I had
learnedabout controllingblood sugar.
After a year of premed courses and another year of waiting, I en
tered the Albert Einstein College of Medicine in 1979.1 was forty-five
years old. Duringmy first year of medical schoolI wrotemy first book,
Diabetes: The GlucografMethod forNormalizing Blood Sugar, enumer
atingthe full details of my treatment fortype 1,or insulin-dependent,
In 19831 finally openedmy own medicalpractice nearmy home in
Mamaroneck, New York. By that time, I had well outlived the life ex
pectancyof an"ordinary" type 1diabetic. Now, by sharing my simple
observations, I was convinced I was in a position to help both type 1
and type 2 diabetics who stillhadthe best years of their lives ahead of
them. I could help others take control of their diabetes as I had mine,
and live long, healthy, fruitful lives.
The goal of this book is to share the techniques and treatments I
have taught my patients and used on myself, including the very latest
developments. If you or alovedone suffers from diabetes, I hope this
book will giveyou the toolsto turn your lifearoundasI did mine.
I wouldliketo thank the following people, whose aidandguidance
made this book possible:
FrankVinicor, MD, MPH, past president of the American Dia
betes Association, who took time from hisoverwhelming schedule to
write the foreword.
Stephen Stark, novelist, critic, andessayist, whose suggestions about
tone, clarity, and structure wereof immeasurablevalue.
Pharmacists Stephen Freed andDavid Joffee, whowrotethe impor
tant appendix"DrugsThat MayAffect BloodGlucose Levels." Patricia
A. Gian, dearfriendand director of my medicaloffice,who shared the
stresses of this endeavor and gave me invaluable aidand guidance all
along theway. Two top-of-the-line professionals, Elizabeth Nagle, my
editor, and Channa Taub, my literary agent, whoseefforts made this
undertaking possible. Peggy Leith Anderson, copyeditor nonpareil.
Karen A. WeinstockandTimothy J. Aubert, for the recipes.
Finally, my loveandthanks to my wife, Professor Anne E. Bernstein,
MD, FAPA, FABPN, who allowed me to steal somuch time that really
belongedto her.
Dr. Bernstein's
Diabetes Solution
Before and After
You're the only person whocan be responsible for normalizing
your blood sugars. Although your physician mayguide you,the
ultimate responsibility isin your hands. This task will require
significant changes in lifestyle that may involve some sacrifice. The
question naturally arises, "Isit really worththe effort?" As you will see
in this chapter, others have already answered this question for them
selves. Perhaps theirexperiences will give youthe incentive to find out
whether you can reap similar benefits.
Thomas G. Watkins is aforty-year-old journalist. His diabetes was diag
nosed twenty-three years ago. For the past nine years he's been following
one ofthe treatment protocob described in this book for people who re
quire insulin.
"Following the instructions of several diabetologists over a period
of years, I hadthe illness 'under control' At least that's what theytold
me. After all, I was taking two shots aday, and adjusting my insulin
doses depending on urine test results, and later on bloodsugar mea
surements. I was also following the common recommendation that
carbohydrates fill at least 60 percent of my caloric intake.
"But something was not right; my life was not 'relatively normal'
enough. I was avoiding heavyexercise for fear of mybloodsugar drop
ping too low. My meal schedulewas inflexible. I still had to eat break
fast, lunch, anddinner even whenI wasn't hungry. Aware that recent
research seemed to associate highbloodsugars with an increased risk
of long-term complications, I triedto keepblood sugars normal, but
wound up seesawing daily between lows and highs. By the end of
4 Before andAfter
1986,1had ballooned to 189 pounds and was at a loss for how to lose
weight. My'good control' regimen hadleft me feeling out of control.
Clearly, somethinghadto be done.
"In that year, I attended ameeting of medical writers at which Dr.
Bernstein spoke. It became clear that his credentials were impressive.
He himself at that time had lived with the disease for four decades and
was nearly free of complications. His approach hadbeen formulated
largely through self-experimentation. His knowledge of the medical
literature was encyclopedic. Some of his proposals were heretical; he
attacked the usual dietary recommendation and challenged dogma
surrounding such basics as howinsulin ought to be injected. But it
seemed likehewas doing something right. During histalk, I hadto use
the bathroom twice; he didn't.*
"I decided to spend aday athisoffice to gather material for anarti
cle tobepublished intheMedical Tribune. There, hisindependence of
thought became clear. 'Brittle' diabetes [entailing anendless sequence
of widebloodsugar fluctuations] was amisnomer that usually indi
cated aninadequate treatment plan or poor training, morethan any
inherent physical deficit, he said. Normal blood sugars round-the-
clock were not just an elusive goal but were frequently achievable, if
the diabetic hadbeentaught the proper techniques. Beyond treatment
goals, he armed his patients with straightforward methods to attain
them. His secret: small doses of medication resulted in small mistakes
that were easilycorrectable.
"Bythen, myinterest had become more personal than journalistic.
Inearly 1987, still wary, I decided to give it atry. The first thingI no
ticedwas that this doctor visitwas unlikeanyprevious ones.Most had
lasted about 15 minutes. This took 8 hours. Others said I had no com
plications; Dr. Bernstein found several. Most said my blood sugars
were just fine; Dr. Bernstein recommended I make changes to flatten
them out and to lower my weight. Those hours were spent detailing
the intricacies involved in controlling blood sugar. His whole ap
proach blastedthetheory espousedbymy first doctor — thatI should
depend on himto dole out whatever information I needed. Dr. Bern
stein made it clear that for diabetics to control their disease they
needed to know as much as their doctors did about the disease.
Very highbloodsugars cause frequent urination.
Beforeand After 5
"Two arguments commonly rendered against tight-control regi
mens arethat they increase the incidence oflowblood sugar reactions
andthat they cause subjects to gain weight. I have foundthe opposite
to be true: I shed about 9 pounds within four months aftermy first
visit, and, years later, I have kept them off. And oncethe guesswork of
how much to inject was replaced by simple calculations, my blood
sugarlevelsbecame more predictable.
"For the first time since I was diagnosed, I felt truly in control. I no
longer amat the mercy of wide moodswings that mirror wideswings
in blood sugar. Though I remain dependent on insulin and all the
paraphernalia that accompany its use, I feel more independent than
ever. I amcomfortable traveling to isolated areas of the world, spend
ing an hour scuba diving, or hikingin the wilderness, without fear of
beingsidetracked by diabetes. Nowif I feel likeskipping breakfast, or
lunch, or dinner, I do so without hesitation.
"I nolonger have delayed stomach-emptying, which can cause very
lowblood sugars right after ameal followed byhighblood sugars many
hours later. My cardiac neuropathy, which is associated with an in
creased risk for early death, hasreversed. Though I eat more fat and
proteinthan before, my blood lipidshaveimprovedand are now well
within normal ranges. My glycosylated hemoglobin measurements,
used bylife insurance companies todetect diabetics among applicants,
wouldno longer give me away. Most important, I now feel well.
"Many doctors will not embraceDr. Bernstein's work, for the sim
ple reason thatDr. Bernstein demands acommitment of time, energy,
andknowledge not only from patients, but from physicians. Diabetics
are the bread and butter of many practices. For decades, the usual
treatment scenario hasbeen abloodtest, ashort interview, a prescrip
tion for a one-month supply of needles, a handshake, and a bill. But
thatis changing. Inthe past few years, evidence has been amassing in
support of Dr. Bernstein's modusoperandi. Nolonger isthe oldhigh-
carbohydrate diet unquestioned; moreandmoredoctors are espous
ing a multiple-shot regimen controlled by the patients themselves.
Most important, though, tight control is beingassociated with fewer
of the diabetic complications that can ravage every major organ sys
tem in the body. Dr. Bernstein's scheme provided me with the tools
not onlyto obtain normal bloodsugars, but toregain afeeling of con
trol I had not had sincebefore I was diagnosed."
6 Before andAfter
Frank Purcell isaseventy-six-year-old retiree who, like many ofmymar
ried patients, works closely with his wife to keep his diabetes on track.
Eileen, who goes bythe nickname Ike, tells the first part ofhis story.
Ike: "Frankhad been treated for many years for diabetes, and had
beentreated orally because hewas atype 2.As far as wewere aware, he
had a functioning pancreas. The thing was, as a younger man, he'd
been told that he hadhighblood sugar, but it was ignored. This was
going back tohisarmy days, in 1953 orso. Noonesuggested medica
tion, no one called it diabetes, and nothing more wasdone. They just
said he had high blood sugar. They called it 'chemical' diabetes. It
showed up on bloodtests, but not on urinalysis. I guess in thosedays,
having it show up on a urinalysis was some sort of determinant. He
didmodifyhisdiet— he stopped eating somuchcandy, andhe took
off weight — helostabout 30 pounds in those days.
"In about 1983, Frank had a mild heart attack. He began to see a
cardiologist, who has been monitoring his health care verycarefully
since then. For about two to three years, he took beta blockers and
maybe oneortwoheart medications. As far as wecould tell, hisheart
problems were verymuchinresolution — I mean he'dhadaheart at
tack, he'dhad no surgery, andseemed to be doingokay. But when he
started working withthecardiologist, the doctor notedthat hisblood
sugar thingwas ongoing, and he began to feel it was of concern. He
prescribed Diabinese, which was the oral medication of choice of the
time, I guess, and hemonitored Frank's blood sugar about every four
"I might say that I never even knewwhat anormal bloodsugar was.
No one ever talked about it. I had no idea whether it was 1,000 or 12.
The onlything wewere ever toldwas that it was high orwasn't high.
This went on andon forcloseto sevenor eight years. If he had seenDr.
Bernstein back then, who knows what could have been different? But
eventually, the cardiologist said he thought Frank ought to see anen
docrinologist. He didn't feel hewas able tocontrol Frank's blood sugar
well enough himself with medication, and so he felt the condition
warranted closer attention.
"Wewent to see agentleman whowas chiefof the diabetes clinic at
amajor hospital here in upstate NewYork, where welive. Now, thisis
averywell thoughtof medical facility. Thedoctor met withus,andhe
kept Frank onthe Diabinese, and monitored himevery three months
or so. Hisbloodsugars were 253,240, andhe wouldsay, 'Let's try an
other pill.' It was always medication. Glyburide, Glucophage — the
Before andAfter 7
wholebit. But tryingto gethisbloodsugar down wasverydifficult. No
one evermentioned diet, really. And rarely wasit everbelow200when
we went in. Rarely. When I finally found out what the numbers meant,
I said to the doctor, 'Don't you think we ought to see a dietitian? I
mean, we'reeatingthe same foodwe always have.' We wereon the nor
mal diet that anybody'son. I had friends who are diabetics who watch
certain things that they eat,and soI thought it made acertain amount
of sense. Hesaid, 'Sure. That's areally goodidea.'
"He gave us the name of a youngwoman, and we saw her three
times. She said, 'Eat eleven carbohydrates every day,' and she gave
us the food pyramid —we didn't need her for that — and nothing
changed, except Frank stopped eating dessert. He would have the oc
casional bowl of icecream, or a piece of cake when he felt like it, or
a cookie. I always bought the newest foods that came out — low-
fat, low-sugar. I was more concerned about fat during that stage, as I
"This went on until God intervened. I mean that. What happened
was, Frank had anattack of serious hypoglycemia [low blood sugar].
No one had warned us that this could happen. No one had told us
what hypoglycemia looked like. I thought it was astroke. Hewas out
of hishead. He couldn't answer questions. The onlything that gave me
some smidgen of doubt was that he got up andwalked to the bath
room and put on histrousers. I called 911. Whenthe medic got here,
hehooked himuptosome glucose, puthimonagurney and trundled
him out of here, and headed for the medical center. In the middle of
the ride, Frank woke upand said, 'What the hell am I doing here?' The
young man said he certainly seemed to be coming out of his stroke
well. Bythetimewegot tothehospital, hewas virtuallyhimself. When
they decided to doa finger stick, his blood sugar was 26, 26 mg/dl. I
didn't have the education in diabetes that I've gotten with Dr. Bern
stein, but I knewenough toknowthat this was not good. Who knows
what it wasbeforehe got the intravenous?
"Now, we'll never knowif heaccidentally took his oral medication
twice thenightbefore — it's very possible —but I tell you, however it
happened, it was the Lord who was watching over Frank and said,
'Nowit's time todo something.' As scary as itwas, itwas also ablessing.
"I have adoctor friend who's aclose colleague of Dick Bernstein's.
My friend hadhadanuncle who'd beenveryill with diabetes andits
complications, but hislife hadbeen prolonged in amuch morecom
fortable fashion by Dick Bernstein. I would talk to my friend about
8 BeforeandAfter
Frank's diabetes, andhe'dsayto me, 'Nothing'sreally goingto change.
You're not going to get hisblood sugars down untilyousee DickBern
stein.' Even though my friend is a doctor, I brushed off his advice.
Frank was seeing a doctor. Why would some private doctor be any
more capable than thehead of thediabetes clinic at amajor medical
center? But after this episode with hypoglycemia, Frank went to my
friend's office with me, and my friend laid it out for him, told us in
grinding detail what we could expect from Dr. Bernstein, what it
wouldbelike, andhowhehopedwewouldrelate to Dick, because he's
rather controversial, andhowhardit was going to be — howmuch of
acommitmentit was going to take. Wewent away thinking, 'Let's give
it atry.'"
Frank: "To be honest, when I first met Dr. Bernstein, I felt he was
somewhat of a flake. I hadworked with doctors in the army, and I was
used to aparticular kind of guy. Dr. Bernstein —now, he's ahorse of
another color. Until I came across him, I never met a doctor who was
so focused on onething. He issocompletely directed toward this one
failing of the human body that I kind of thought that maybe it was a
little too intense. But the results have been rather spectacular, and I'm
very happy with him. He has specific programs, he has direction, he
has goals, and heisnot sidetracked byanything other than tending to
diabetes. He's given mearegimen. I keep track of myblood sugar, and
it's pretty much under control. Instead of blood sugar counts of over
200,1 nowget theminthe range of 85 to 105, whichwas thegoal heset
for me. I takeinsulin in the morning andbefore middayandevening
meals, and before I go tobed. I don't eat ice cream, and I don't doalot
ofthings Iused todo routinely. When I first came toDr. B., Iwas look
ingvery pale and wan, and nowI'mlooking much ruddier and health
ier. I'm a little irritated with this constant puncturingof my fingers,
but I just doit automatically now, like second nature.
"WhenI found out Iwas going tohave toinject insulin, I justbroke
down and cried. It was like the final straw, and I thought, 'My life is
over.' Now I hardly thinkabout it. I use Dr. Bernstein's painless injec
tion method andit doesn'tbother me at all. It onlytakesasplit second.
Theneedle issotiny, I can barely feel theshots of minute doses of in
sulin. I usethe'lovehandles' on the sides of mywaist. Now, I'm apretty
skinny guy, so there isn't much there, but I can hardly feel it. He made
me do it in the office. He showed me — did it to himself— and then
he made me do it. Since then, I just do it routinely, all on my own. If
Before andAfter 9
I'mout, I do it wherever I am— at atable in arestaurant, in the men's
room, et cetera — I'm not the least bit ashamed and no one seems
much to notice."
Ike: "About theinsulin, I had thefeeling that it was going tobein
evitable, and when Frank got the news he just broke into tears and
really felt that this was the final insult. He'd had many physical prob
lems, and insulin seemed like avery low blow for him. But he didit,
stayed with the program, and within amonth tosix weeks, we began
to feel that we were on top of this, knew what was going on. He can
manage his blood sugar when it's alittle low, when it's alittle high. He
knows just what to do. His overall health has improved since thebe
ginning. Dr. Bernstein really gave us aneducation."
Joan Delaney is afifty-three-year-old mother and financial editor. Her
story is not unusual.
"I must admit that the prospect of following this newregimen for
diabetes control seemed daunting atfirst. My life, Ithought, would be
dominated by needles, testing, and confusion. However, after a few
weeks, the program became a simple part of my day's routine, like
putting on makeup.
"Before Ibecame apatient of Dr. Bernstein, Iwas somewhat resigned
tothe probability of suffering complications from diabetes. Although
I tookinsulin, I in no way felt I had control of the disease. I had leg
pains atnight. Myhands and feet tingled. I had gained weight, having
no understanding of the exchange dietmy previous doctor hadthrust
into myhands. Ibecame chronicallydepressed andwas usuallyhungry.
"Now that I follow ablood sugar normalizing program, I know I
amin control of my diabetes, especially whenI see that number nor
mal most of the time onthe glucose meter. Best of all, I feel good, both
physically and emotionally. I am now thin. I eat healthful, satisfying
meals andamnever hungry. Myleg pains have disappeared, as hasthe
tinglingin myhands and feet. And now that I am in control of the dis
ease, I no longer find the need to hide from friendsthe fact that I have
About65percent of diabetic men are unable to have sexual intercourse,
because high blood sugars have impaired the mechanisms involved inat-
1o Before andAfter
taining erection ofthe penis. Frequently partial, albeit inadequate, erec
tions are still possible; such "borderline" men maystill beable to enjoy
adequate erections for intercourse, after extended periods of normal
bloodsugars. We have seen such improvements in anumber ofpatients —
but only in those whose problem was caused mainly by neuropathy
(nerve damage), as opposed to blockages ofthe blood vessels that supply
the penis. When we initially sawL.D„ in the pre-Viagra era, he asked me
to evaluate his erectile dysfunction. I found that the blood pressures
in his penis and his feet were normal, butthat the nerve reflexes in the
pelvic region were grossly impaired. L.D.'s comments refer in part tothis
"I'm a fifty-nine-year-old male, married, with three children. Ap
proximately four years ago, after being afflicted with type 2 diabetes
foraboutten years, I noticed that I was always tired. In addition, I was
quite irritable, short-tempered, and haddifficulty maintaining con
centration for extended periods of time. Otherwise I was feeling well,
with the exception that I was becoming impotent, having difficulty
maintaining an erection during sexual intercourse. At the time, I had
no knowledge whether these conditions were interrelated.
"AfterDr. Bernstein taught me to measure my blood sugars, I dis
covered that they averaged about 375 mg/dl, whichis veryhigh.With
my newdiet andsmall doses of insulin, they are nowessentially nor
mal all the time.
"I began to feel betterthanI hadin years, both physically andmen
tally. The problem with impotency has improved. I maintain a daily
checkof my blood sugars and feel that my overall improvement has
also helped me recuperate quickly from atotal hip replacement with
out any complications."
RJN, MD, is board certified inorthopedic surgery. Hehas been following
one ofthe regimens described in this bookfor the pastthree years.
"I am fifty-four years old and have had diabetes since the age of
twelve. For thirty-nineyears I hadbeentreated with atraditional diet
and insulin regimen. I developed severe retinopathy, glaucoma, high
blood pressure, andneuropathy that required me to wear alegbrace.
Both of my kidneys ceased functioning, and I was placed on kidney
dialysis for many months until I received a kidney transplant. The
dialysis treatments required me to be in the hospital for about 5hours
Before andAfter 11
pervisit, 3times aweek.Theywere verydebiUtating andleft me totally
"Years of widely fluctuating blood sugars affected my mental and
physical stability, with great injury to my familylife asaresult.The re
sultant disability also forced me to give up my surgical practice, and to
suffer almost total loss of income.
"Frequent lowblood sugars wouldcause me to exhibit bizarre be
havior, sothat people unaware of my diabetes wouldthink I was tak
ingdrugsor alcohol. I washostile, anxious, irritable, or angry, andhad
extreme mood changes. I would experience severe physical reactions
that included fatigue, twitching of limbs, clouding of vision, head
aches, and blunted mental activity. I sufferedmany convulsions from
lowblood sugars andwasplaced in hospital intensive care units.When
myblood sugars werehigh, I hadno energyandwasalways sleepy. My
visionwasblurredand I wasusually thirstyandurinating alot.
"For the past three years, I have been meticulously following the
lessons that Dr. Bernstein taught me. I measure my blood sugars a
number of times eachdayandknowhowto rapidly correct slightvari
ations from my target range. I follow a very low carbohydrate diet,
which makes blood sugarcontrol much easier.
"In return for my conscientious attention to controlling bloodsug
ars, I'vereaped anumber of rewards. Myneuropathyis gone, andI no
longer require a leg brace. My retinopathy, which was deteriorating,
has now actuallyreversed. I no longer suffer from glaucoma, which
had required that I use special eyedrops twiceeach day for more than
ten years. My severedigestive problems have markedly improved. My
mental confusion, depression, and fatigue have resolved so that I am
nowable to work full-time and productively. My blood sugar control
has been excellent.
"I nowdeal with my diabetes in arealistic, organized manner, and
asa result I feel stronger, healthier, happier, and more positiveabout
my life."
J.L.F. isseventy-oneyearsoldandhasthree grandchildren. Hestill works
as a financial consultant, and was a navalaviator in World War II. His
blood sugars are currently controlled by diet, exercise, and pills called
insulin-sensitizing agents. Thanks to thediet described in this book, his
cholesterol/HDL ratio, an index of heart disease risk (seepage 57), has
12 Before andAfter
droppedfrom a veryhigh risk level of7.9 toa below-average level of3.0.
His hemoglobin A1C test, which reflects average blood sugar for the prior
four months, has dropped from 10.1 percent (very high) to 5.6percent
(nearly in the nondiabetic range). His R-R interval study (see Chapter
2), anindicator ofinjury tonerves that control heart rate, has progressed
from aninitial value of9percent variation (very abnormal) toa current
value of33percent, which is normal for hisage.
"I probably had mild diabetes for most of my adult lifewithout re
alizing it. It first appeared as lethargy, later as fainting, stumbling, or
falling, but asrare occurrences. I also had difficultyattainingfull erec
tion of my penis.
"In early 1980,1 beganto experience dizziness, sweating, armpains,
tendencies to fainting, andthe symptoms usually associated with heart
problems. An angiogram revealed severe disease of the arteries that
supplied my heart. I therefore had surgery to open up these arteries.
All waswell for the next sevenyears, and I again enjoyed good health.
"In late 1985,1beganto notice a loss of feeling in my toes. My in
ternist diagnosed it as neuropathyprobably due to high blood sugar.
He did the usual blood test, and my blood sugarwas 400. His advice
wasto watchmy diet, especiallyto avoidsweets. I returned foranother
checkup in 30 days. My blood sugar was 350. Meanwhile, my neu
ropathywas increasing, along with the frequency of visits. My blood
test results were consistently at the 350 level, my feet were growing
more numb, and I was becoming alarmed.
"I felt okay physically, walked at least two milesaday, workedout in
the gym once or twice a week, worked a full schedule as a business
consultant, and didn't worry a great deal about it. But I did begin to
inquireof friends and acquaintances about anyknowledge or experi
encethey might haverelative to neuropathyor diabetes.
"My first jolt camefromastory fromone of my friends who had di
abetes, foot neuropathy, deep nerve painin his feet, and a nonhealing
ulceron atoe. He told me that asthe neuropathyprogressed, amputa
tion of the feet was likely, elaborating by describing the gruesome
'salami surger/ of uncheckeddiabetes.
"That's when I becameemotionally unglued, asthey say. One thing
about aging anddisease, youthink agreat deal about the utter horror
of becoming acripple, dependent upon others for your mobility. Sud
denly foot numbness is no longer acasual matter, more likeahead-on
crash into reality.
"Then I met awealthycardealer at the golf club, with his legscut off
Beforeand After 13
ashighaslegs go, who explained he hadn't paid too much attentionto
his diabetes at the time and his doctor couldn't help him. He could
never leavehis chair, except for relief and sleep, and he had to be lifted
for that. Oh, he was cheerful enough. He joked that they would cut
him off at the middle of his butt the next time, that is, if he didn't die
first. A displayof courageto others was a macabrenightmare to me. I
got serious about getting someone, somewhere, to tell me what to do
about my ever-worseningnumbness, which by now had spreadto my
penis. My condition became an ever-present, gnawing anxiety with
me, acreepingpresenceI couldn't fight against becauseI simply didn't
know howto fight it.
"Then, in earlyApril 1986, mywife and I went to visit Dr. Bernstein.
The first visit lasted llh hours. Each detail of diagnosis and treatment
wasdiscussed. Each symptom of the disease, howeverminute, was de
scribedin great detail, the importance of eachbalancedwith another,
with specific remedies for managing them. Take the seemingly in
significant matter of scaly feet, acommon, dangerous symptom of di
abetes. Dr. B. prescribed mink oil, rubbed into the feet morning and
night. Practiced as directed, instead of split skin and running foot
sores, you have skin as soft and smooth as velvet. Consider the alter
native — feet split, painful, and slow (if at all) to heal — which can
change your entire life. Special shoes, debilitating gait, not to mention
the horrible possibility of progressive amputation; all things that
reallycan happen if your diabetes is not treated properly.
"What is of highest importance, I believe, is the in-depth explana
tion of diabetes, its causes, symptoms, and treatment. He givesyou the
rationale for treatment, so that you have a comprehensive under
standing of what is wrong and how it can be corrected.
"First, through frequent finger-stick blood testing, we came to an
understanding as to the specifics of how to attack my diabetes. We
startedwith diet. It wasn't just eat this, don't eat that, but eat this for
these reasons and eat that for other reasons. Know the reasons and the
differences. Knowingthe how andwhy of diet keeps you on the track,
and the discipline of that knowledge makes control easy. Forwithout
continuous diet observance,you will surelyworsen your diabetes. He
explains that the effect of uncontrolled diabetes on the heart can be
much more deleterious than the other popular demons — choles
terol, fat in the diet, stress, tension, et cetera — demons not to be ig
nored, obviously, but merely put into proper perspectiveto the main
villain — diabetes.
14 BeforeandAfter
"Well, the results for me arethe numbness of my feet and penis have
regressed, and my erections have improved. My feet are now beauti
fully supple and healthy. The severe belching, flatulence, and heart
burn after meals have disappeared. The other ills of diabetes have
apparentlynot greatly affectedme, and now that I know that control
lingmy diabetes is the key to ahealthyheart,I expect to reducegreatly
any future risk of heart attacks.
"One great result of my abilityto normalize my blood sugars has
been the stabilizing of my emotional attitude towardthe disease. I no
longer have asense of helplessness in the face of it; no longer wonder
what to do; no longer feel hopelessly dependent on peoplewho have
no answers to my problems. I feel free to exercise, walkvigorously, en
joy good health without worry, enjoy my precious eyesight without
fear of diabetic blindness, yes, even havea new confidence in normal
sexual activities.
"All of the enjoyments of healththat were slowlyebbing away are
now within my control, and for that I thank my new knowledge and
LeVerne Watkins isa sixty-eight-year-oldgrandmother and associate ex
ecutive director ofasocial service agency. When wefirstmet, she hadbeen
taking insulin for two years, after developing type 2 diabetes thirteen
years earlier. Hercomments relate inpart to the effects of large amounts
of dietary carbohydrate, covered by large amounts of insulin, whileshe
wasfollowing a conventional treatmentplan.
"In less than two years, my weight had increased from 125 to 155
pounds; my appetitewas always ready for the next snack or the next
meal. All my waking hours were focused on eating. I always carried a
bag of goodies — unsalted saltine crackers, regular Coca-Cola, and
glucose tablets. I always hadto eat'on time.' If I wasahalf-hour lateat
mealtime, my hands would begin sweating, I would become very jit
tery, and if in a social gathering or a conferenceor meeting at work, I
would have to force myself to concentrate on what was taking place.
During ameeting that I waschairing, the last thing I remember saying
was,'Oh, I'm so sorry,' before I toppled out of the chairto wake up and
find myself in the emergencyroom of alocal hospital.
"During a subway ridewhich generally took about 25minutes, the
train was delayed for closeto 2 hours and — to my utter dismay — I
had forgotten my bag of goodies. As I felt myself 'going bananas,'
Before andAfter 15
sweating profuselyand perhaps acting a little strange, a man sitting
across fromme recognized my MedicAlert bracelet, grabbed my arm,
and screamed, 'She has diabetes!'
"Food, juice, candy bars, cookies, and fruit came from all direc
tions. It wasacold, wintery day, but peoplefanned and fed me. And I
was so grateful and soveryembarrassed. I stoppedridingthe subway,
and rescheduled as many meetingsand conferences as I could to take
place directly after lunch so that I would have more time before the
next snackor meal would be necessary.
"I felt that I hadno control over my fife; I was constantly eating, I
outgrewall my clothing, shoes and underwear included. I had been a
rather stylish dresser since college days. NowI felt rather frumpy, to
say the least. Once, I triedto discuss with my diabetologist how I was
feeling about gaining weight and eating all the time. I was told, 'You
just don't have any willpower,' and'If you put your mind to it, you
wouldn't eat so much.' I was very, veryangry, somuch so that I never
consultedhim again.
"Onmy own, I triedWeight Watchers, but thediet I hadbeengiven
by the dietitian to whom the diabetologist had referred me did not
mesh with theWeightWatchers diet. So along I limped, trying to ac
cept that I was getting fatter each day, was always hungry, had no
willpower, andmost of the time was feeling unhappy.
"My husbandwas my constant support throughall this. He would
say, 'Youlook goodwith a few more pounds Gobuy yourselfsome
newclothes,' especially when I wouldaskhim to zip somethingthat I
was trying to squeeze into. He always clipped newspaper and maga
zinearticles about diabetes andwouldremindme to watch specials on
TV. He encouraged me to be active in the local diabetes association,
andwould accompany me to lectures and various workshops. Then,
on Sunday, April 3, 1988 — Easter Sunday — he clipped an article
from the New York Times entitled'Diabetic Doctor Offers a NewTreat
ment.' Little did I realize that this thin news article would be a newbe
ginning of my life with diabetes. I must have read it several dozen
times before I finally met with Dr. Bernstein. Since that first meeting,
I haven't had one single episode of hypoglycemia, which I had for
merly experienced veryoften. Following theregimen of correcting my
high and low blood sugars, takingsmall doses and different kinds of
insulin, and eating meals calibrated for specific amounts of carbohy
drates and protein, my outlook brightened andI began to feel more
energetic and more in charge of myself and my life. I couldnow hop
16 BeforeandAfter
on the train, ride the subway, driveseveral hours, and not fear one of
those lowblood sugar episodes. I started once again to exercise every
day. My stamina seemed to increase. I didn't have to pushhardto ac
complish my daily goals at work and at home. Within a couple of
months, I was back to 129 pounds, had gone from size 14to size 10,
and ten months laterto size 8 and 120pounds. Eventhe swellingand
painin my right knee — arthritis, I wastold — abated. I feel great. My
self-esteemand self-worthare whole again. I now take only 8 units of
insulin each day, whereI had previously been taking31 units.
"I amalso conquering my uneasyand frightening feelings about the
long-termconsequences of having diabetes. While I oncethought that
heart disease, kidney failure, blindness, amputations, and many other
healthproblemswerewhat the future probably held for me, I nowbe
lievethat they are not necessarily outcomes of livingwith diabetes.
"But my life is not perfect. I stilloccasionally throwcautionto the
wind by eating too much andeating foods I know are taboo. Sticking
with my diet of no bread, no fruit, no pasta, no milk, seemed easy
when it wasnew,but now it isnot easy, andloadsof my efforts gointo
making salads, meat, fish, or poultryinteresting andvaried. My fan
tasies are almostalways of some forbidden food — ahot fudge sundae
with nuts, or my mother's blueberrycobbler topped with homemade
ice cream. But when all is told, I feel that I am really lucky. All my ef
forts have really paidoff."
AD. is a fifty-five-year-old former typesetter whose diabetes was diag
nosed fourteen years ago. As with many other people who use our regi
men, his test ofaverage blood sugar (hemoglobin A1C) and his tests for
cardiac disease risk (cholesterol/HDL ratio) simultaneously droppedfrom
high levels toessentially normal values.
"I watched my mother deteriorate in front of me fromthe compli
cations of diabetes, finally resulting in anamputationof the legabove
the knee, and a sorrowful existenceuntil death claimedher. My oldest
brother, who was also diabetic, was plagued with circulatory compli
cations that resulted in the amputation of both feet, with unsightly
stumps. Diabetes robbedhim of anormal existence.
"When I began to experience the all-too-familiar diabetes symp
toms, my future looked bleak andI feared the same fate. I immediately
searched for help, but for two years floundered around gettingmuch
Before andAfter 17
medical advice but not improving. In fact, I was getting sicker. My
doctorhad said, 'Watch your weight,' and prescribed asingledailyoral
hypoglycemic pill for my type 2 diabetes. It sounded easy, but it wasn't
working. My glucoselevelswere in the 200rangeall too often, and oc
casionally reached400.1 was constantly exhausted.
"I startedDr. Bernstein's programin 1985. Sincethen I have recov
eredmy former vitality and zest for life. At my first visit, he switched
me to another approach — a fast-acting blood sugar-lowering pill 3
times aday, beforemeals, along with aslower-acting pill in the morn
ing and at bedtime. My regimen was totally overhauled to eliminate
foods that raised blood sugar, and to reduce greatlymy consumption
of carbohydrates in general. Macaroni andravioli hadbeen important
parts of my diet since birth. I had to give these up. I didn't mind a
greater emphasis on protein. I even began to include fresh fish in
my diet.
"My initialreaction wasthat theserestrictions weretoo high a price
to pay, and that I would be unableto continue them for long. Also, I
wasaskedto check myblood with ablood sugarmeter for aweek prior
to every visit to Dr. Bernstein. That meant stickingmy finger several
times a day. I waswillingto discipline myself for ashort period in or
der to be able to return to a more active, vigorous life and to put my
malaise to rest. At the beach, I wassorely tempted to giveup the diet,
while watching family and friends eat without restrictions. But since
my body was feeling healthier, I continued with the program. After
about two months, with many dietary slips on my part, I managedto
better discipline myself because I sensedit made me feel better. My
glucose level startedto descend to 140,130, and finally to 100 or less
on a consistent basis.
"Dr. Bernstein also encouraged me to purchase a pedometer, a de
vice that clipped to my belt and measured the distance that I walked
each day. I began to walkdaily, holding3-poundweights andswinging
my arms. This was yet another thing to bother with, and I felt it would
cut into my free time. But the resultwas an invigorating high. By this
time, I didn't mind pricking my fingers several times each day, as it
showed me the way to better blood sugars. Fortunately for me, New
Rochelle hasmany beautiful parks. I choseGlenIsland Park because it
is nearLong Island Sound andnicelykept. This meant gettingup ear
lierin the morning to walkduringthe week,but that was no problem
since I am an earlyriser. I bought some cast-irondumbbells for addi-
18 Before andAfter
tional exercise. I learned about arm curls, overhead raises, arm circles,
and chest pulls. I didn't realize that there were so many different exer
cisesthat you could do at home to benefit your health.
"My glucoselevels arenow consistentlywithin or near the normal
range,not at the sorrylevels which nearlyput me in the hospital. That
all-consuming fatigue is gone, and I feel that now I'm in control of my
diabetesinsteadofthe reverse. With adherence to the program, I know
that I don't have to suffer the same debilitating effects that afflict so
many other diabetics."
Harvey Kent isfifty-one. He has known abouthisdiabetes for approxi
matelysixyears, andwesuspect that heprobably hadit for three tofour
years priorto his diagnosis. Hehas afamily history of diabetes, and his
storyisfairly typical.
"I went in for aroutine physical. I've always had high risk factors —
both my parents had diabetes, my brother had diabetes, and my sister
has diabetes. My brother, who was forty-nine, passed away recently
from diabetic complications. My sister, who is fifty-nine, is on dialysis.
When I found out I had it, I felt I was going down the same slippery
slope. I'd been trying to lose weight, but not very successfully. The
doctor I was seeing, an endocrinologist, kept upping my medication.
Everytime I went to seehim, I wound up taking more and more, and
my blood sugars weren't goinganywherebut up.
"I kept havingthe feeling that as far astreatment went, nothing was
happening. I wasn't in bad shape, but then I watched my brother pass
away, and I thought, 'I've got to do something.'
"I happen to live in Mamaroneck, New York, near Dr. Bernstein,
andmywife suggested that I seehim forasecondopinion. I kept won
dering, 'Is there another approach?' That's really how it started. The
standard approachwas always to tell me to lose weight, to exercise,and
to take medication. I wastrying to do allthose things, but I wasn't hav
ing much success at anyof them exceptthe takingof medication. As it
turned out, Dr. Bernstein still saidthe same three things, but his ap
proach to each of the categories wasradical, especially on the diet. The
diet has been a major factor — I've lost alot of weight.
"Once I started getting a senseof what Dr. B. was talking about —
which was reallyright from the firstvisit; he's very thorough in his ex
planations— I kind of figured it out. Just to demonstrate the effects of
diet, he told me to stayon my same diet and measure my blood sugars,
BeforeandAfter 19
but I startedcutting back on the carbohydrates, so by the time we sat
down to negotiate a meal plan, which was maybe the third or fourth
session, he just confirmed what I'd already started about a month
"Before I met Dr. Bernstein, I'd been under treatment for diabetes
by three different doctors.The guy I was seeing before Dr. B. is an en-
docrinologist/diabetes doctor with a fairly large practice. He never
oncesaid to me, 'Youknow,by controlling yourblood sugars, most of
these complications arereversible.' When Dr. B. told me that — well,
for a diabetic who's stuck with this disease for the rest of his life, that's
niceto hear. Nobodyever tells youthis. At least I don't rememberany
one everexplainingthis to me. I'vebeen amember ofthe ADA [Amer
ican Diabetes Association] for several years, and no one ever said
anythinglikethat to me, anywhere. I waslucky. I hadn't developed that
many complications — not like my brother and sister— but I knew
how fast they could get you.
"With my olddoctor, I'dbeentoldto monitormy bloodsugars and
then comein everythreemonths. What it wassupposedto do, I wasn't
sure— keep you honest, maybe, but I couldn't figure that out. I was
checking my fasting blood sugars in the mornings. They wereaverag
ing somewhere about 140 mg/dl. And when I'd go in, the doctor
would do blood work, scratch the bottoms of my feet, and check my
eyes, then say, 'See me in three months.' The whole thing would take
maybe half an hour and then I'd see him again in three months. I
wasn't sure what the whole thing was about. The thing is— and I
found this out with my sister and my brother— it's a slippery slope.
You start out as a type 2 and you get this kind of treatment, and you
burn out your pancreas, andbeforelong, you're insulin-dependent.
"When I sawDr. B., he did a very extensivemedical exam and un
covered everything there was to uncover. He checkedeverything. He
found that I had an anemia, and so we started doing things to deal
with that. I had not had retinopathyor neuropathy. I had some protein
in my urine, a potential sign of kidney disease. But he saidthat could
be frommy old kidney stone, or it couldbe fromthe diabetes. He said
we'd wait awhile until my blood sugars were normalized, then test
again and find out, because if it wasthe diabetes, it should clear up.
"The first thing he did was get me off Micronase and onto Glu-
cophage. Micronase is one of those oral hypoglycemic agents that
stimulate your pancreas, and he said, 'Why are you doing this? You're
burning your pancreas out quick.' Helookedat my blood sugars care-
20 Beforeand After
fullyand told me I waslowat particular times of the dayand told me
what I had to do to cover the valleys as well as the peaks. Insulin. I
never wanted to take insulin. My father did it, and the idea just
brought back horrible memories. My other doctor would say, 'All else
is failing, now you have to go on insulin.' What Bernstein says is, 'I
want you to takeinsulin in orderto coveryour peaksand to keep your
pancreas from burning out.' This seems to me a much more sensible
"My wife is very perceptive about the whole thing, and she said
what I really needed was a coach, and Bernstein is very much like a
coach. Having read up about him and knowing that he was an engi
neer, you can seethe differencein his approach.You can seeless of the
medical model and more of an engineering model: he's putting you
back together, taking your components and manipulating them in or
der to accomplish something. He's a diabetic himself, he knows the
thing insideand out, andsoyou get the sense that he'smuch more ac
tively involved. Now I measuremy blood sugars 5 times a day, but in
stead of just jotting them down and saying come back in three
months, he adjusts the medication, using it to tweak the peaks and val
leys, to get the most optimum response. Now I haveexcellentcontrol.
"The diet takes some getting used to. Most diabetics, I would sur
mise, love to eat. Especially if you come from a culture where food is
the coin of the realm. People ask me now, 'What do you eat?' I say, 'I
have turkey, some salad, and a Diet Coke.' I used to be a big pancake
eater. Talk about your carbohydrate! Every Saturday and Sunday
morning for years I would make pancakes for my wife. Now I make
them for her and for my daughter and don't have any — or occasion
allystealjust abite — and I miss it, but I am so much more in control
now, and I feel so much better. I've seen so much of my family go
down the slippery slope,it seems a small sacrifice for good health.
"Since the time I started seeing Dr. Bernstein, I've lost close to 30
pounds. My blood sugars have dropped by about 35 percent, but
my weight loss was not on a weight loss diet, just on Dr. Bernstein's
meal plan. I still haveawayto go,but forthe firsttime I feel like I'm in
J.A.K. isasixty-seven-year-old business executive whohadhadtype2 di
abetes for twenty-four years, and had been taking insulin for twenty,
when hestartedonourregimen. Hewrites thefollowing:
Beforeand After 21
"I visited Dr. Bernstein on the recommendation of some good
friends, asI had just lost the central vision in my right eye due to sub-
retinal bleeding.
"It took hours of instruction, counseling, and explanationto make
me clearly understand the relationships between diet, blood sugar
control,and physical well-being. I washoping for the possibilitythat I
might experience animprovement in my already deteriorated physical
condition. I havediligently followed up on what I wastaught, and the
results are obvious:
• I no longerhavecrampsin my calves and toes.
• The neuropathy in my feet has normalized.
• Variousskin conditions havecleared up.
• Tests for autonomic neuropathy(R-Rintervalstudy) totally nor
malized in only two years.
• The difficulty I had with digestion hascleared up completely.
• My weight dropped from 188 to 172 pounds in six months.
• My original cholesterol/HDLratioof 5.3put me at increasedrisk
for a heart attack. With a low-carbohydrate diet and improved
blood sugars, this value has dropped to 3.2, which puts me at a
lowercardiac risk than most nondiabetics of my age.
• My daily insulin dose has dropped from 52 units to 31 units, and
I no longer have frequent episodes of severe hypoglycemia.
• My overall physical condition and stamina have improved con
"All these improvements occurred because I learned how to control
my blood sugars. As a matter of fact, my glycosylated hemoglobin (a
test that correlates with average blood sugar during the prior four
months) dropped from 7.1 percent to 4.6 percent, so that I am now in
the samerange asnondiabetics. I havedeveloped full confidencein my
abilityto managemy own diabetes. I understandwhat is happening. I
can adjust and compensate my medications asthe need arises.
"If I haveto miss a meal, for whatever reason, I canadjust accord
inglyandamnot tiedto aclock, asI wasbeforeI learned thesenewap
proaches to blood sugarcontrol.
"I would saythat not only hasmy physical condition improved, but
my mental attitude is far better today than it was ten or fifteen years
ago. My only regret is that I did not learn howto be in charge of my di
abetes yearsearlier."
22 Before andAfter
Lorraine Candido has had type 1 diabetes for more than twenty years
and has been mypatient for ten. She is in her sixties, andsheand her
husband, Lou, her "copilot," work together tokeep her blood sugars nor
mal. Like a lotof happily married couples, Lorraine andLou sometimes
almost speak asone. When Lorraine comes infortreatment, Lou is with
her. When shecalls onthephone, Lou isonthe other line. They talk about
howstarting theprogram changed their lives:
Lorraine: "I had alot of complications. Bladderinfections, kidney
infections — and then my eyes.My feetwerenumb up to my heels. As
a matter of fact, one day I waswalkingbarefoot and I wasn't aware of
it but I had a thumbtack in my foot all daylong. I had neuropathy of
the vagusnerve. I had an ulcer from medication. My mother had had
eye problems,and sowhen I went to anophthalmologist,he said, 'You
have some of your mother's problems.We'll keep an eye on you; come
back in a year.' And I thought, 'Uh-oh.' When Dr. Bernsteinexamined
my eyes, he said, 'Oh, I'll make an appointment for you.' Right away I
had laser surgery."
Lou: "I firmly believe that if she hadn't gone to Dr. Bernstein, she
would've been blind. Her last two visits to the eye doctor she got ex
cellent reports. As a matter of fact, he saidhe had no idea where the
fluid in one eye had gone, but it was all gone."
Lorraine: "I was elated. He said my left eye had made great
progress and I was doing well.
"When I first met Dr. Bernstein, I had no idea what I was getting
into. All I knew wasthat I wasn't feeling well and I was goingnowhere.
I was kind of scared, didn't know what I was getting into, and didn't
know if I wanted to. It was plain and simple. I liked Snickers candy
bars. He said,'No.' I couldn't haveanything I liked and wanted, and we
kind of butted heads— but then I realized, 'Hey, come on, is there
reallya candy bar worth dying for?'
"He's avery gentle gentleman. I think he's extremely caring; you're
not treated like cattle, you're treated as a person, and he answers all
your questions. Between the two of us, at the beginningwehadalot of
questions. Really, I don't know if I couldlivewithout him.
"We found him — it's kind of embarrassing, but our son used to
have a newsstand, and Lou would go help him out on Sundays, and
Lou would bring me home the papers to read. Well, in one of those
horrible tabloids — you know, when they run out of weird stuff, they
run unusual medical stories reprinted from somewhere else — the
headline on this was'Diabetic Heals Himself,' and you know, we didn't
Before andAfter 23
think that much about it. But I wasn't feeling well, and so we made
some inquiries. Nowof course we're in adifferent state and nobody I
knew had ever heard of him, but we called his office. I didn't talk to a
nurse or someone, he got on the phone himself and he offered us ref
erences. Well, that settled it rightthere. I mean,howmany doctors do
youknowof who'd offer youreferences? So Lousaid, 'Pack up, honey,
we're going.'"
Lou: "Shehad a doctor up here in Springfield, Massachusetts, she
was seeing and I was getting pretty concerned about it. Her feet were
getting numb, she had kidney problems. I don't have diabetes, but I
happenedto havethe samedoctor asmy internist, and I said to him,
'Isn't there something you can do for my wife?' He had a son who
worked at the Joslin Clinic, which we hadheard was very good. 'Can
we take her to the Joslin Clinic?' But he said, 'What can he do for her
up there that we can't do for her here?' We got sort of scared. They
were running her the standard way they treat diabetics — standard
but safe. Safe for them, but not much help for Lorraine.
"At Dr. Bernstein's, to start, it was a 10-hour training period— two
5-hour sessions that she had to take at the start."
Lorraine: "It was my husband, me, and the doctor. No waiting
room for hours. Now, to be honest, when we walked out of there —
it's a 2-hour drive between our house and there — I didn't want to do
it. But on the drive back home after the first session, we talked. We
talked constantly, and I knew I didn't want to do it, but I also knew I
was goingto do it. Common sensejust dictated it. I wanted to live, and
I wantedboth feet and both eyes. It was plain and simple. The feeling
in my feet has come back almost 100 percent, by the way."
Lou: "We found out about the diet on the first visit, and it took
about a month to get her blood sugars into the target range. She had
beenrunning 300,400mg/dl bloodsugars prettyregularly."
Lorraine: "I was kind ofreluctant to start with. It was clear that Dr.
Bernstein's program wasn't a ride in an amusement park. In some re
spects,it was awhole new way of living, and we had to changeall our
grocery lists — but I had a supportive friend here in Lou. When I
startedon the diet, we pretty much atethe same food. He didn't have
to, but he did. He would have a few extras here and there and I
wouldn't, but it was years before I could go into the supermarket, be
causeit felt like I couldn't have anything there. It was very hard to get
used to. I resented being told what to do and how to do it."
Lou: "It's very difficult. You have to understand something. When
24 Before andAfter
she started the programshe was closeto sixty years old, and we were
accustomedto livingin a particular way."
Lorraine: "We have grandkids — we've been married forty-five
years — wehave sixkids andseven grandkids, andthey comeover for
chocolate chip cookies and icecream."
Lou:"The programworks—"
Lorraine: "Because I'm still here."
Lou:"—but it's difficult to do, becauseyou reallyhave to be dedi
Lorraine: "Let's put it this way. There are no hot fudge sundaes
here. Ever. Not for Thanksgiving, not for Christmas, birthdays, an
niversaries— there are no deviations from the program. The first
week, because of the change in diet, I lost 15 pounds. You looked at
what you wereeating, measuredit —"
Lou: "It was a combination of things. The amount of insulin
changedalot. Shewastakingsometimes 80to 90units of insulin on a
dailybasis, and nowshe's taking 13^2 units. Insulinis the fat-building
hormone, so reducing your dosage changes things substantially. And
you'rechanging the amount of carbohydrate you'retaking in, and so
she lost all this weight."
Lorraine: "Altogether, I lost 85 pounds. I wear junior size clothes.
Call me stubborn, but I still resent being told what to eat."
Lou: "Let me put it this way. You five a quality of life and give up
what you have to —"
Lorraine: "Like fudge."
Lou: "Or potatoes. The point is, you have to decide somewhere
along the line. Are you going to live and enjoy the rest of your life
without problems,or are you goingto fight the realityof the situation
and go down the tubes? It'sachoice."
Lorraine: "It's an attitude. I don't like his program, but it works. I'm
still here. I miss the goodiesI givemy grandkids, allthe cookies, candy
bars,ice cream. And the holidays. Everything's kind of restricted."
Lou: "The irony of this is, my wife, sinceshe lost allthe weight, she
dresses in very sporty clothes. Now, I'm aracewalker. She doesn't exer
cise, but because ofheredityorwhatever, shehasbeautiful,stronglegs,
and so she wears these spandex tights and such, and people ask her,
'How much do you run?'"
Lorraine: "He's a champion racewalker, very self-disciplined. Not
me. I had a conversation with God, and He said, 'Don't sweat.' I'm
Lou's cheerleader. I stayhome and readbooks."
Before andAfter 25
Lou: "She walks withme sometimes. But I laugh my ass off."
Lorraine: "It's fun to go shopping andbuy junior sizes with my
granddaughters — but I don't let them borrow my clothes. Before I
started the program, I neverthought about howI looked, how I felt —
allI know is, the clothesI wasbuyingwereone size fits all."
Lou: "Now look at her."
By the way, Lorraine's cholesterol/HDL ratio has droppedfrom ahigh
cardiac risk 5.9 toa verylowrisk 3.3.
It isn't unusual for people with diabetes tomake major changes in other
aspects of their lives once their blood sugars have been restored to normal
after years ofpoor control. The changes that we see include marriages,
pregnancies, andreentry into the workforce. The story of Elaine L. falls
into the last category. She also points outthe disablingfatigue thatshe ex
perienced when her blood sugars were high. This problem has led other
diabetics, desperate to retain their abilities tofunction productively, to
abuse amphetamines. Elaine is a sixty-year-old mother and artist. Her
story is not unusual.
"When I developed diabetes twenty-one years ago, I began a fruit
lessodysseyto learnall I couldabout this disease andto havethe tools
to be able to deal with the psychological and physical roller coaster
that I was experiencing.
"The hardest thing to cope with was the total loss of control over
my life. I was told that I was a'brittle' diabetic and that I would just
haveto endure the very high and very lowblood sugars that were to
tally exhausting me. I feared that my eyeswould be damaged. I'm an
artist, and this frightened me the most. I knew that this diseasewas de
stroyingmy body every day and that I was helpless.
"We went from doctor to doctor and to major diabetes centers
around the country. I never could get a handle on how to become
'controlled.' I was given a gold star for'good' blood sugarby one doc
tor; told I 'had imbued the number 150with mystical significance' by
another; informed that if my blood sugars werehighafterlunch today,
I could correct them before lunch tomorrow. All the while, I was feel
ing worse and worse. I stopped painting. I was just too tired. I was so
scared to read any more of the diabetes magazines, because I kept
learning more and more about what was in store for me.
"I'd been diabeticabout five years when an uncle in Florida advised
me to read Dr. Bernstein's first book. It made a lot of sense, but when
26 Before andAfter
I read it, I thought, 'Diabetes has robbed me of so much already, I
don'thave anymore timeoreffort togive toit — andwhowants tobe
aprofessional diabetic?' Of course, there was alot of anger anddenial
andevenattempts to forget about being diabetic. Maybe I couldforget
about it for awhile, but it never forgot about me.
"Aseedwas nowplanted, however, in spiteofmyself. I knewthat no
matter whathappened down theroad, I needed to feel that I hadtried
everything possible, so that I would never have to say, 'I wish I had
done more.'
"I was very wary of my first visit to Dr. Bernstein's office. I really
thought I would hate having tochange mydiet yetagain. I didnot rel
ishthe idea of multiple daily injections, testing my bloodsooften,and
keeping records. The fact isthatI didhate all of thatuntil I found I was
recording better andbetter bloodsugars. Thediet wasn't anymorere
strictive than the American Diabetes Association diet I had been fol
lowing, andmost important, I was feeling better andmuchless tired.
In fact, I began to paint again and soon rented a studio. I now paint
full-time, but this time I actually sell my work.
"The regimen that I feared has, in the end, given me the freedom of
which I had dreamed."
Although Elaine does notmention it inher story, her cholesterol/HDL
ratio dropped from an elevated cardiac risk level of4.74 to the "cardio
protective" level of 3.4, as her long-term blood sugars approached nor
mal. Furthermore, her weight has dropped from 143 pounds to 134
pounds, and her hemoglobin A1C has dropped from avery high 10.7per
cent to a nearly normal 6.0percent.
Carmine DeLuca is inhis early sixties andhas hadtype2 diabetes since
about ageforty-five. Like many ofmypatients, hehadbeen in"standard"
treatmentandfound hiscondition gettingprogressively worse.
"I wastakingpills, triedsomediet changes, but afterabout ten years
my diabetes just gotworse. Through the years, as adiabetic, I hadseen
some articles about Dr. Bernstein, and he had appeared several times
in the local newspaper. A colleague at work mentioned this Dr. Bern
steinto me, the sameguywho hadbeenin the paper. Shesaid, 'If you
ever want to go to someone, go to this guy' And I heard from a few
other peoplearoundthe area who said, 'He'sexcellent.'
"Over the years, I've had trouble with my eyes, my feet, and my
hands, but that was before Dr. Bernstein saw me. I had tried to watch
Before andAfter 27
my diet, but beingItalian, youknow, you're always involved with the
pasta, thebread, andsoforth, andsoI really didn't doverywell on di
eting. Apparently the pill that I was taking was hterally burning me
out. I was just going to a general doctor, an internist, and what did he
know? I usedto keepblood sugar about 140 to 160, and then all of a
sudden it started hittingthe200 mark, andit was starting tohit it con
sistently, and then closeto 300, and then over 300, and the nerve end
ings in my feet were gone, andthe feeling in my hands. I did have, at
age fifty, two cataracts. I don't knowif you want to blameit on dia
betes, but I guess you can. Finally, when it was sohigh, I said, 'Well,
something hasto be done. What have I gotto lose?'
"And so when thetimecame, Ithought, let mego tothebest. Every
bodytalks about howexcellent he is, soI made anappointment. My
bloodsugars were veryhigh,in the high300s, like375. When I sawDr.
Bernstein, I hadnoidea whatI was getting myselfinto. I hadjust heard
that he was one of the best, and so I said, 'Lemme do it.' He struck me
as very, very knowledgeable. I learned an awful lot — he told me
things about diabetes that I just never heard about, even from people
with diabetes. He made you feel good, because he Hterally grew up
with it. He was very professional, yet you couldsit down and talk to
him. He said he was always available, available 24hours aday, andhe
has been,no matterwhat. Yougointothat, andyou feel prettygood.
"I've lost weight since I started seeing him. A few pounds hereand
there, but the thingis, even though I haven't taken off alot of weight
yet, everybody says, 'Hey, you look great.' But you could see, priorto
seeing Dr. Bernstein, that it was tearing me down, people couldseeI
wasn't looking that good.
"Starting the program was tough, but it was carbohydrates that
were killing me. He put me on the diet. I never had a problemwith
cholesterol, but for some reason, every time youturn around, people
are talking about high cholesterol this, high cholesterol that, so I
thought about it. But I didn't give a damn about carbohydrates; no
body talks about carbohydrates and cholesterol. At least until Dr.
Bernstein said,'You don't eat this, you don't eat that,'and I said,'These
are all carbohydrates.' And soI'mon the diet and,boom, I startlosing
a little weight.
"The thingwasto get usedto doingwithout the carbohydrate, but
it'sokay, because I like meat, I likesalad, I likevegetables. I caneat all
the cheese I want — I mean, within reason. My blood sugar hasbeen
good, averagingunder 100, and I feel like a million.
28 Before andAfter
"I'mstrictly on insulin and onepill, and we've reduced the insulin,
andas my bloodsugar improves, I think we'll reduce it evenmore. I
see him nowevery twomonthsor so, and for aweekprior, I measure
my sugars 4timesaday andbring thechart to him. He really analyzes
it — youknow, 'All right, take this, don't dothis. We'll reduce this. Don't
eatthat.' He's got asystem all hisownandit'sgreat. It works. It canbe
apain in the neck, but hey. He tells me I'magood patient. I'mhere to
prove thatit's not impossible to change, and theresults are there."
Mark Wade, MD, is one of many physicians with diabetes. He is board
certified inpediatric medicine. His lovely wife notlong ago gave birth to
their third child. His story has a number ofparallels with my own.
"Dr. Bernstein's program turned my life around. Prior to meeting
Dick Bernstein at age thirty-four, I hadspent twenty-twoyears of my
fife as what I then considered a well-controlled insulin-dependent,
juvenile-onset diabetic. I'd never been hospitalized for ketoacidosis [a
serious condition caused by high blood sugar in combination with
dehydration] or severe hypoglycemia, hadwhat I considered good cir
culation and nerve function, exercised daily, and atepretty much what
ever I felt like eating.
"However,cuts andlacerations took months or years to heal instead
of days, andalways leftugly scars. Once ortwice each year, I would de
veloppneumonia that typically lasted four months andhadme, with
out fail, out of school or work for two and a half months per episode.
My mood swings went from kind andlovable to short-tempered, hot
headed, and uncaringfourto five times daily, congruent with my rou
tine blood sugar swings from high blood sugars (300 to 500) after
mealsto hypoglycemia (less than50)beforemeals. This Dr. Jekyll/Mr.
Hyde personality made me very unpredictable and unpleasant to be
around, and came closeto causingme to lose my wife and the close
nessof family and friends. I was forced to eat my meals at exactly the
same times each dayin order to avoid life-threatening episodes of low
blood sugar. Even so, I hadto adjust my lifearoundthe inevitable pe
riods of hypoglycemia. If I didn't eat, my lifewas in trouble, and un
fortunately sowerethe people who hadto interactwith me when I was
hypoglycemic. Most of the times thosewerethe ones I lovedmost.
"My trainingasaphysician, asaninternandresident, averaging 110
hours aweek of work, wasat times anightmare, though I did it, trying
to balance rounds, clinics, emergencyroom and ICUschedules, screen-
Before andAfter 29
ing patients, longhours of reading, andanunreal demand on physical
tolerance, emotionalstability, andconsistency that almost drove me to
the breaking point. My mission was to be an excellent doctor, and I
was, with acalm, cooldemeanor whichI presented externally. But in
side I was a mess, and my interactions with my loved ones and close
friends were horrible. I was an avid basketball player, jogger, and
weight lifter, but despite doing these activities daily, I found my perfor
mance and endurance were usually modulated by my blood sugar —
and was never really surewhether I would be able to perform for 10
minutes or 2 hours. In addition, despite my highlevel of exercise, 1to
\Vihours daily for twelve years, I was never able to develop amuscu
lar or athletic body type, eventhoughI worked hardat it.
"I was always extremely conscientious about testing andexercising
andeating anddoctor visits, tothepointthatmy friends thoughtI was
neurotic. I wasconsistently following the conventional guidelines rec
ommended to diabetics, and I thought I was a rather model patient.
The problems that I described above, I had been led to believe, were a
natural part of lifefor adiabetic. Nooneshowed me that my lifecould
bebetter, that I could control my diabetes rather thanlet my diabetes
control me,thatwithrecognition of afew principles thatare really just
common sense, a few extra finger sticks anda few extrainjections and
better control of my dietary intake — I could bein charge for real!
"Nine years ago, I met Dick Bernstein. Dr. Bernstein not only gave
me the most complete, comprehensive, logical, reasonable, and infor
mative teaching on diabetes that I have ever encountered, but his
uniquely expert andcomprehensive physical examination andtesting
illuminated for me the most accurate picture of my overall health and
thesubtle tolls thatthe previous management of my diabetes hadper
mitted. Then with a personalized, comprehensive, tightly controlled
but reasonable diet, exercise, and a bloodsugar-monitoring plan, he
put me in control of my diabetes for the first time. Sure, the diet plan,
finger sticks, and 5 to 8 painless insulin injections a day for my pro
gram require a high degree of discipline and self-control, but it's
doable, it works, andthis comparatively small sacrifice brings me the
freedom of lifestyle, quality of life, and longevity thatnondiabetics take
for granted.
"The results have been as follows: I can eat or fast whenever I
choose. I plan my day around my activities rather than around my
meals, have the ability to be much more flexible in my schedule
and participation in activities, andnowhave the ability to adjust my
30 BeforeandAfter
daily activities easily to accommodate'emergencies' or suddenchanges
in schedule— activities and adjustments that nondiabetics take for
granted. Best of all, the wildmood swings have beeneliminated and
I'm sick much less often and lessseriously."
All of these people have been patients of mine andhave seen wonderful
improvements in their health. Ifyou're curious about howpeople have
fared using the prior two editions ofthis book, I urge you to look at the
testimonials on the Web sitefor this book at www.diabetes-book.com/
testimonials/testimonials.shtml and those in reader reviews of theprior
editions on www.amazon.com to see similar reactions from people
who have tried the program but have never been under my direct
care. For somereason, readers intheUnited Kingdom presentmore com
plete and more impassioned reviews. Many of these can be seen at
www.amazon.co.uk. It is well worth a visit. These people havebeen suc
cessful inspite of the major obstacles imposed by their National Health
Before You Start
Diabetes is so common in this country that it touches nearly
everyone's fife — orwill. The statistics on diabetes are stag
gering, and a diagnosis can be frightening: diabetes is the
third leading cause of death in the United States. According to the
most recent statistics compiled by the National Institutes of Health
(NIH), as of 2005, a staggering 7 percent of the U.S. population, or
nearly 21 million people, have diabetes, with 14.6 million diagnosed
and6.2 millionwhohave not yetbeendiagnosed. This number willno
doubt increase. Most death certificates of diabetics do not fist diabetes
asthe underlying cause of their heart attacks, strokes, or fatal infec
tions. If it were included, it mightwell betheleading cause of death in
the United States. Recent reports predict that 95 percent of people
borntoday in theUnited States will eventually develop diabetes.
Even more alarming, the incidence of type 2— or what was once
known as maturity-onset diabetes — among children eighteen years
old and younger has skyrocketed. A Yale University study of obese
children between ages four andeighteen appeared in the March 14,
2002, issue of the New England Journal ofMedicine. The study found
that nearly a quarter hada condition that's oftena precursor to dia
betes. According to USA Today's story on the report the same day,
"The incidence of type 2 diabetes, the form that usually occurs in
adults, has increased in young people, especially Hispanics, blacks,
andNative Americans. Some regional studies suggest theincidence of
type2in children has jumped from less than 5%, before 1994, to upto
50%." That children are increasingly getting a disease that once tar
geted fifty- to sixty-year-olds presents anewand frightening potential
public health disaster.
34 BeforeYou Start
Each year, tens of thousands of Americans lose their eyesight be
cause of diabetes, the leading cause of newblindness for peopleages
twenty-five to seventy-four. Ninety-five percent of diabetics have
type2diabetes. Because 80 percent of type2diabetics are overweight,
manyinappropriately feel thatthedisease istheir own fault, the result
of some failure of character.
Since youare reading this book, youor aloved onemayhave been
diagnosed recently withdiabetes. Perhaps youhave long-standing di
abetes and are not satisfied with treatment that has left you plagued
with complications such as encroaching blindness, foot pain, frozen
shoulder, inabilityto achieve or maintaina penileerection, restrictive
lungdisease, hip andleg pain, or heart or kidneydisease.
Although diabetes is still an incurable, chronic disease, it is very
treatable, and the long-term "complications" are fully preventable.
For morethansixtyyears, I'vehadtype 1diabetes, also called juvenile-
onset or insulin-dependent diabetes mellitus (IDDM). This form of
diabetes is generally far more serious than type 2, or non-insulin-
dependent diabetes mellitus (NIDDM), although both have the po
tential to be fatal.* Most type 1 diabetics who were diagnosed back
about the same time I was are nowdead from one or more of the seri
ous complications of the disease. Yet after living with diabetes for
more than sixty years, instead of being bedridden or out sick from
work (or dead, the most likely scenario), I am more fit than many
nondiabetics who are considerably younger than 1.1 regularly work
12-hour days, travel, sail, and pursue avigorous exercise routine.
I am not special in this regard. If I can take control of my disease,
you can take control of yours.
In the next several pages I'll give you ageneral overviewof diabetes,
how the body's system for controlling blood sugar (glucose) works in
the nondiabetic, and howit works — and doesn't work — for diabet-
*Fora periodof time,manypeople considered thedesignations type 1and type2
out of date, replacing them with the terms IDDM and NIDDM, which are
slighdymisleading and are losingcredence. Whileit is true that most of those
with type 2 can stayalive without injecting insulin, many patients who suffer
from type 2, or so-called NIDDM,do inject insulinto preservetheir health. The
terms"autoimmune diabetes" for type 1and"insulin-resistantdiabetes"for type 2
are more precise, but are unlikely to takeoverfor the much-easier-to-say type 1
and type 2. The situation is further complicated by the recent discovery that
most type 2 diabetesalso has an autoimmune component.
Diabetes: TheBasics 35
ics. In subsequent chapters we'll discuss diet,exercise, and medication,
and howyou can use them to controlyour diabetes. If discussion of
diet and exercise sounds like "the same old thing" you've heard again
and again, read on, because you'll find that what I've observed is al
most exactlytheopposite of"the same oldthing," which iswhat you've
probably been taught. The tricks you'll learn can help you arrest the
diabetic complications you may now be suffering, may reverse many
of them, andshould prevent theonset of newones. We'll also explore
new medical treatments andnewdrugs that are now available to help
manage bloodsugar levels andcurtail obesity.
Diabetes is the breakdown or partial breakdown of one of the more
important of the body's autonomic(self-regulating) mechanisms, and
its breakdown throws manyother self-regulating systems into imbal
ance. Thereisprobablynot a tissue in thebodythat escapes the effects
of the high blood sugars of diabetes. People with high blood sugars
tend to have osteoporosis, or fragile bones; they tend to have tight
skin; they tendtohave inflammation andtightness at theirjoints; they
tend to have manyother complications that affect every part of their
body, including the brain, with impaired short-term memory and
even depression.
Insulin: What It Is, What It Does
At the centerof diabetes is the pancreas, a large glandabout the size of
your hand, whichis located towardthe backof the abdominal cavity
and is responsible for manufacturing, storing, and releasing the hor
mone insulin. The pancreas also makes several other hormones, as
well as digestive enzymes. Even if you don't know much about dia
betes,in all likelihoodyou'veheard of insulin and probablyknowthat
we all have to have insulin to survive. What you might not realize is
that only a small percentageof diabetics must haveinsulin shots.
Insulinis a hormone producedbythe beta cells of the pancreas.In
sulin's major function is to regulate the level of glucose in the blood
stream, which it does primarily by facilitating the transport of blood
glucose into most of the billions of cells that make up the body. The
presenceof insulin stimulates glucosetransporters to move to the sur
face of cells to facilitate glucose entry into the cells. Insulin alsostimu-
36 BeforeYou Start
lates centers in the hypothalamus of the brain responsible for hunger
and satiety. Indeed, there is some insulin production even as one be
gins to eat, before glucose hits thebloodstream. Insulin also instructs
fat cells to convert glucose and fatty acids from the blood into fat,
which the fat cells then store until needed. Insulin is an anabolic hor
mone, whichis to saythat it is essential for the growth of many tissues
and organs.* In excess, it can cause excessive growth—as, for exam
ple, of body fat and of cells that line blood vessels. Finally, insulin
helps to regulate, or counterregulate, the balance of certainother hor
mones in the body. Moreabout those later.
One of the ways insulinmaintainsthe narrowrangeof normal lev
els of glucose in thebloodisbyregulation of theliver andmuscles, di
recting them to manufactureand store glycogen, a starchysubstance
the body uses when blood sugar falls too low. If blood sugar does fall
even slightlytoo low— as mayoccur after strenuous exercise or fast
ing— the alpha cells of the pancreas release glucagon, another hor
mone involved in the regulation of bloodsugarlevels. Glucagonsignals
the muscles and liver to convert their stored glycogen back into glu
cose (a processcalledglycogenolysis), whichraises blood sugar.When
the body's stores of glucose and glycogen have been exhausted, the
liver, and to a lesser extent the kidneys and small intestines, can trans
form some of the body's protein stores — muscle mass and vital or
gans — into glucose.
Insulin and Type 1 Diabetes
As recently as eighty-five years ago, before the clinical availabiHty of
insulin, the diagnosis of type 1 diabetes —which involves a severely
diminished or absent capacityto produce insulin — was a death sen
tence.Most peoplediedwithina fewmonths of diagnosis. Without in
sulin, glucose accumulates in the blood to extremelyhigh toxiclevels;
yet since it cannot be utilized by the cells, many cell types will starve.
Absent or lowered fasting (basal) levels of insulin also lead the liver,
kidneys, and intestines to performgluconeogenesis, turning the body's
protein store — the muscles and vital organs— into even more glu
cose that the body cannot utilize.Meanwhile,the kidneys, the filters of
the blood, try to rid the body of inappropriately high levels of sugar.
*Anabolicand catabolichormones normallyworkin harmony,buildingup and
breaking down tissues, respectively.
Diabetes:TheBasics 37
Frequent urination causes insatiable thirst and dehydration. Eventu
ally, thestarving bodyturns more and more protein to sugar.
The ancient Greeks described diabetes as a disease that causes the
bodyto melt into sugar water. When tissues cannot utilize glucose,
theywill metabolize fat for energy, generating by-products called ke
tones, which are toxic at very highlevels and cause further waterloss
asthe kidneys try to eliminatethem (see the discussion of ketoacido
sis and hyperosmolar coma, inChapter 21, "How toCope withDehy
dration, Dehydrating Illness, and Infection").
Today type1diabetes isstill avery serious disease, andstill eventually
fatal if not properly treated withinsulin. It can kill yourapidly when
yourblood glucose level is too low— throughimpaired judgment or
loss of consciousness while driving, for example — or it can kill you
slowly, byheart orkidneydisease, whichare commonlyassociatedwith
long-term bloodsugar elevation. Until I broughtmy bloodsugars un
der control, I had numerous automobile accidents due to hypogly
cemia, and it's only through sheer luck that I'm hereto talk about it.
The causes of type 1diabetes havenot yet been fullyunraveled. Re
search indicates that it's an autoimmune disorderin which the body's
immune system attacksthe pancreatic beta cells that produce insulin.
Whatever causes type 1 diabetes, its deleterious effects can absolutely
be prevented. The earlier it's diagnosed, and the earlier blood sugars
arenormalized, the better off you will be.
At the time they arediagnosed, many type 1diabetics still produce
a small amount of insulin. It's important to recognizethat if they are
treated early enough and treated properly, what's left of their insulin-
producing capabilityfrequently canbe preserved. Type 1 diabetes typi
cally occurs before the age of forty-five and usually makes itself
apparentquite suddenly,with suchsymptoms asdramaticweight loss
and frequent thirst and urination. We nowknow, however,that assud
den asits appearance may be, its onset is actually quite slow. Routine
commercial laboratory studies are available that can detect it earlier,
and it may be possible to arrest it in these early stages by aggressive
treatment. My own body no longer produces any detectable insulin
at all. The high blood sugars I experienced during my first year with
diabetes burned out, or exhausted, the ability of my pancreas to pro
duce insuhn. I must have insulin shots or I will rapidly die. I firmly
believe— and know from experiencewith my patients — that if the
kind of diet and medical regimen I prescribe for my patients had been
utilized when I was diagnosed,the insulin-producing capabilityleft to
38 BeforeYou Start
me at diagnosis would likely have been preserved. My requirements
for injected insulinwouldhave beenlessened, andit wouldhavebeen
much easier for me to keep my blood sugars normal.
Blood Sugar Normalization: Restoring the Balance
According to the NIH, approximately 225,000 people died in 2002
from diabetes, but it is likely that deaths due to diabetes are under-
reported. (Is a diabetic's death from heart disease, kidney disease, or
stroke, for example, really adeath fromdiabetes?) It is the NIH's con
tention that "the risk for death among people with diabetes is about
twice that of peoplewithout diabetes of asimilarage."
Certainlyeveryone hasto dieof something,but you needn't die the
slow, torturous death of diabetic complications, which often include
blindness and amputations. My history and that of my patients sup
port this.
The Diabetes Control and Complication Trial (DCCT), conducted
by the NIH's National Institute of Diabetes and Digestive and Kidney
Diseases (NIDDK), beganin 1983 asaten-yearstudy of type 1diabet
ics to gauge the effects of improvedcontrol of blood sugar levels. Pa
tients whose blood sugars were nearly "normalized" (my patients'
blood sugars are usually closer to normal than were those in the in
tensivecare armof the trial because of our low-carbohydrate diet) had
dramatic reductions of long-term complications. Researchers began
the DCCT trying to seeif they could, forexample, lessenthe frequency
of diabetic retinopathy by at least 33.5 percent.
Instead of a one-third reduction in retinopathy, they found more
than a75percent reduction inthe progression of early retinopathy. They
found similarlydramatic results in other diabetic complications and
announced the results of the study early in order to make the good
news immediately available to all. They found a 50 percent reduction
of risk for kidney disease, a 60 percent reduction of risk for nerve
damage, and a35 percent reduction of risk for cardiovascular disease.
This reduction continues to this day, many years after the study was
I believethat with truly normal blood sugars, which many of my
patients have,these reductionscanbe 100 percent.
The patients followed in the DCCT averaged twenty-seven years of
ageat the beginning of the trial, so reductions could easilyhave been
greater in areas such ascardiovascular disease if they hadbeen older or
followed for a longer periodof time. The implication is that full nor-
Diabetes: TheBasics 39
malization of blood sugar could totally prevent these complications.
Inanycase, the results of the DCCT aregood reason to begin aggres
sively to monitor and normalize blood sugar levels. The effort and
dollar cost ofdoing sodoes nothave toberemotely as high as was sug
gested in the DCCT'sfindings.
The Insulin-Resistant Diabetic: Type 2
Different from type 1 diabetes is what is officially known as type 2.
This isbyfarthe moreprevalent form of thedisease. According to sta
tisticsfrom the American Diabetes Association, 90-95 percent of dia
betics are type 2. Furthermore, as many as a quarter of Americans
between the ages of sixty-five and seventy-four have type2diabetes. A
study published by Yale University found that 25 percent of obese
teenagers now have type 2 diabetes.
(A new category of "pre-diabetes" has been recently called latent
autoimmune diabetes, or LADA. This category applies to mild dia
betes with onset after the age of thirty-five, in which the patient has
been found to produce an antibody to the pancreatic beta cell protein
called GADA, just as in type 1 diabetes. Eventually these people may
developovert diabetes and require insulin. When the symptoms of di
abetes finallyoccur, they are often more severe than at the "onset" of
type 1 diabetes.)
Approximately 80 percent of those with type 2 diabetes are over
weight and are affectedby a particular form of obesityvariously known
as abdominal, truncal, or visceral obesity. It is quite possible that the
20 percent of the so-called type 2 diabetics who do not have visceral
obesityactuallysuffer from a mild form of type 1 diabetes that causes
only partial loss of the pancreatic beta cells that produce insulin.* If
this proves to be the case, then fullyall of those who have true type 2
diabetes may be overweight. (Obesity is usually defined as being at
least 20 percent over the ideal body weight for one's height, build,
and sex.)
Whilethe causeof type 1diabetes maystillbe somewhatmysterious,
the causeof type 2 is less so. As noted previously, another designation
for type 2diabetes is insulin-resistant diabetes. Obesity, particularlyvis
ceral obesity, and insulin resistance —the inability to fully utilize the
glucose-transporting effects of insulin—are interlinked. For reasons
* Recentstudies showthat even type 2 diabeticsexperiencesome degree of im
mune attack on their beta cells.
Hereditary Craving
for Carbohydrate Foods
High Dietary
Fig. 1-1. The vicious circle of insulin
by Reduced
Number of
Beta Cells
Beta Ceil Burnout
BeforeYou Start
related to genetics (see Chapter 12, "Weight Loss— If You're Over
weight"), asubstantial portionofthe populationhasthe potentialwhen
overweight to become sufficiently insulin-resistant that the increased
demands on the pancreas burn out the betacells that produceinsulin.
These peopleenterthe viciouscircle depictedin Figure 1-1.Note in the
figure the crucial roleof dietarycarbohydrate in the development and
progression of this disease. This is discussed in detailin Chapter 12.
Insulin resistance appears to be caused at least in part by inheri
tance and in part by high levels of fat — in the form of triglycerides
released from abdominal fat — in the branch of the bloodstream that
feeds the liver. (Transient insulin resistance can be created in labora-
Diabetes: TheBasics 41
toryanimals byinjectingtriglycerides — fat — directlyintotheir liver's
blood supply.)* Abdominal fat is associated withsystemic inflamma
tion, another cause of insulin resistance, as areinfections. Insulin resis
tance by its very nature increases the body's need for insulin, which
therefore causes the pancreas to work harder to produce elevated in
sulin levels (hyperinsulinemia), which can indirectly cause high blood
pressure anddamage the circulatory system. High levels of insulin in
theblood down-regulate theaffinity for insulin thatinsulin receptors
alloverthe body havenaturally. This"tolerance" to insulincauses even
greaterinsulin resistance.
So, to simplifysomewhat, inheritance plus inflammation plusfat in
the blood feeding the liver causes insulin resistance, which causes ele
vated serum insulin levels, which cause the fat cells to build even more
abdominal fat, which raises triglycerides in the fiver's blood supply
and enhances inflammation, which causes insulin levels to increase
because of increased resistance to insulin.
If that sounds circular, it is. But note that the fat that is the culprit
here is not dietary fat.
Triglycerides are in circulation at some level in the bloodstream at
alltimes. High triglyceride levels are not so much the result of intake
of dietary fat as they are of carbohydrate consumption and existing
body fat. (We will discuss carbohydrates, fat, and insulin resistance
more in Chapter 9, "The Basic Food Groups") The culprit is actually
a particular kind of body fat. Visceral obesity is a type of obesity in
which a special kind of fat is concentratedaround the middle of the
body, particularly surrounding the intestines (the viscera).A man who
is viscerallyobese has awaist of greater circumference than his hips. A
woman who is viscerallyobese will have a waist at least 80 percent as
big around asher hips. All obese individualsand especially those with
visceralobesity areinsulin-resistant. The ones who eventually become
diabetic arethose who cannot makeenough extrainsulin to keep their
blood sugars normal.
Though treatment has many similar elements — and many of the
adverse effectsof elevated blood sugar are the same— type 2 diabetes
differs from type 1in several important ways.
The onset of type 2 diabetes is slowerand more stealthy, but even in
its earliest stages the abnormal blood sugar levels, though not sky-
* New evidence demonstrates a role for fat contained in muscle cells (intramy-
ocyte fat) as another important factor in causing insulin resistance.
42 Before You Start
high, can cause damage tonerves, bloodvessels, heart, eyes, andmore.
Type 2diabetes isoften called the silent killer, and it isquite frequently
discovered through oneof itscomplications, such ashypertension, vi
sual changes, or recurrent infection.*
Type 2 diabetes is, at the beginning, a less serious disease —pa
tients don't melt away into sugarwaterand diein a few months' time.
Type 2, however, can through chronically but less dramatically ele
vated blood sugars be much more insidious. Because so many more
people areaffected, it probably causes more heartattacks, strokes, and
amputationsthan the more serious type 1 disease. Type 2 is a major
cause of hypertension, heart disease, kidney failure, blindness, and
erectile dysfunction. That these serious complications of type 2 dia
betescanprogress is no doubt because it is initiallymilder and is often
left untreated or treated more poorly.
Individuals with type 2 still make insulin, and most will never re
quire injected insulin to survive, though if the disease is treated
poorly, they can eventually burn out their pancreatic beta cells and
require insulin shots. Because of their resistance to the blood sugar-
lowering effects of insulin (though not its fat-buildingeffects), many
overweight type 2diabetics actually makemore insulin than slimnon
Since high blood sugar is the hallmark of diabetes, and the cause of
everylong-term complicationof the disease, it makes sense to discuss
where blood sugar comes from and howit is used and not used.
Our dietary sources of blood sugar are carbohydrates and proteins.
One reason the taste of sugar — a simple form of carbohydrate — de
lightsus is that it fosters productionof neurotransmitters (principally
serotonin) in the brain that relieve anxiety and can create a sense of
well-beingor even euphoria. This makescarbohydrate quite addictive
to certain people whose brains mayhaveinadequate levels of or sensi
tivity to these neurotransmitters, the chemical messengers with which
the brain communicates with itself and the rest of the body. When
* A common early sign of mild chronic blood sugar elevation in women is re
current vaginalyeastinfections that causeitchingor burning.
Diabetes: TheBasics 43
blood sugar levels are low, the liver, kidneys, and intestines can,
through aprocess we will discuss shortly, convert proteins into glucose,
butveryslowlyandinefficiently. The bodycannot convert glucose back
into protein, nor can it convert fat into sugar. Fat cells, however, with
the helpof insulin, do transform glucose into saturatedfat.
The taste of protein doesn't excite us as much as that of carbohy
drate —it would bethevery unusual child who'd jumpup anddown
in the grocery store and beg his mother for steak or fish instead of
cookies. Dietary protein gives us a much slower and smaller blood
sugar effect,which, as you will see,we diabetics can use to our advan
tagein normalizingblood sugars.
The Nondlabetic
In the fasting nondiabetic, and evenin most type 2 diabetics, the pan
creasconstantlyreleases a steady, lowlevel of insulin. This baseline, or
basal, insulin level prevents the fiver, kidneys, and intestines from in
appropriately converting bodily proteins (muscle, vital organs) into
glucose and thereby raising blood sugar, a process known as gluco-
neogenesis. The nondiabetic ordinarily maintains blood sugar im
maculatelywithin a narrowrange — usuallybetween80and 100mg/dl
(milligrams per deciliter),* with most peoplehoveringnear 85 mg/dl.
There are times when that range can briefly stretch up or down — as
high as 160mg/dl and as lowas 65—but generally, for the nondia
betic, such swings are rare.
Youwill note that in some literature on diabetes, "normal" may be
defined as 60-120 mg/dl, or evenas high as 140mg/dl. This "normal"
is entirely relative. No nondiabetic will haveblood sugar levelsas high
as 140mg/dl except after consuming a lot of carbohydrate. "Normal"
in this case has more to do with what is considered "cost-effective" for
the average physicianto treat. Since a postmeal (postprandial) blood
sugar under 140mg/dl is not classified as diabetes, and since the indi
vidual who experiences such a valuewill usuallystill have adequate in
sulin production eventually to bring it down to reasonable levels,
manyphysicians would seeno reasonfor spending their valuabletime
on treatment. Suchan individual maybe sent off with the admonition
to watch his weight or her sugar intake. Despite the designation "nor-
* A deciliter is one-tenth of a liter, or a little over 3 ounces. A milligram is one
one-thousandth ofa gram, or about one three-thousandth of the weight of sugar
in a levelteaspoon.
44 Before You Start
mal,"anindividual frequendydisplaying ablood sugar of 140mg/dl is
a goodcandidate for full-blown type 2 diabetes. I have seen"nondia
betics"with sustainedblood sugars averaging 120 mg/dl develop dia
betic complications.
Let's take a look at how the average nondiabeticbody makes and
uses insulin. Suppose that Jane, a nondiabetic, arises in the morning
and has amixedbreakfast,that is, one that contains both carbohydrate
and protein. On the carbohydrate side, she has toast with jellyand a
glass of orange juice; on the protein side, she has a boiled egg. Her
basal (i.e., before-meals) insulin secretion has kept her blood sugar
steadyduring the night, inhibiting gluconeogenesis. Shortly after the
sugar in the juiceor jellyhits her mouth, or the starchy carbohydrates
in the toast reach certain enzymes in her saliva and intestines, glucose
beginsto enterher bloodstream. The mere presence of food in her gut
as well as the rise in Jane's blood sugarsignal her pancreas to release
the granules of insulin it has storedin order to offset a jump in blood
sugar (see Figure 1-2). This rapid release of stored insulin is called
phase I insulin response. It quicklycorrects the initialblood sugar in
crease and can prevent further increase from the ingested carbohy
drate. As the pancreas runs out of stored insulin, it manufactures
more, but it has to do so from scratch. The insulin released now is
known as the phase II insulin response, and it's secreted much more
slowly. As Jane eats her boiled egg, the small amount of insulin of
phase II can cover the glucose that, over a period of hours, is slowly
produced from the protein of the egg.
Insulin acts in the nondiabetic as the means to admit glucose—
fuel — into the cells. It does this by activating the movement of glu
cose"transporters" within the cells. These specialized protein mole
cules protrude from the cytoplasmof the cellsand their outer surfaces
to grab glucose from the blood and bring it to the interiors of the
cells. Once inside the cells, glucose can be utilized to power energy-
requiringfunctions. Without insulin, the cells can absorb only avery
smallamount of glucose, not enoughto sustain the body.
As glucose continues to enter Jane's blood, and the beta cellsin her
pancreas continue to release insulin,some of her blood sugar istrans
formed to glycogen, a starchy substance stored in the muscles and
liver. Once glycogenstoragesites in the muscles and liver are filled, ex
cess glucose remaining in the bloodstream is converted to and stored
as saturated fat. Later, as lunchtime nears but before Jane eats, if her
blood sugar drops slightlylow, the alpha cells of her pancreas will re-
Diabetes: The Basics
• Phase 1
Phase II
'<- Baseline
- ^- Baseline
' Start of Meal
Digestion Finished
Fig. 1-2.Phase I andphaseII insulin
response ina normal, nondiabeticperson.
Terry Eppndge
lease anotherpancreatic hormone, glucagon, which will "instruct"her
liver and muscles to begin converting glycogen to glucose, to raise
bloodsugar. When sheeats again, herstore of glycogen will bereplen
Thispattern of basal, phaseI, then phaseII insulinsecretion is per
fect for keeping Jane's bloodglucose levels in a safe range. Her bodyis
nourished, and things work according to design. Her mixed meal is
handledbeautifully. This is not, however, howthings work for either
the type 1or type 2 diabetic.
The Type 1 Diabetic
Let's look at what would happen to me, a type 1 diabetic, if I had the
same breakfast as Jane, our nondiabetic.
UnlikeJane, becauseof a condition peculiar to diabetics, if I take a
long-acting insulin at bedtime, I might awaken with a normal blood
sugar, but if I spendsometimeawake before breakfast, mybloodsugar
mayrise, evenif I haven't had anythingto eat. Ordinarily, the liver is
constantlyremovingsome insulin from the bloodstream, but during
the first fewhours after waking from a full night's sleep, it clears in
sulinout of the bloodat an accelerated rate. Thisdip in the level of my
previouslyinjectedinsuliniscalled the dawnphenomenon (seeChap
ter 6,"StrangeBiology"). Because of it, mybloodglucose can riseeven
though I haven't eaten. Anondiabeticjust makesmore insulin to off
set the increasedinsulin clearance. Those of us who are severely dia
betic have to track the dawn phenomenon carefully by monitoring
46 BeforeYou Start
blood glucose levels, and canlearn howto use injectedinsulin to pre
vent its effect upon blood sugar.
As with Jane, the minute the meal hits my mouth, the enzymes in
my saliva beginto breakdown the sugars in the toast andjuice,and al
most immediately my blood sugar would begin to rise. Even if the
toast had no jelly, the enzymes in my saliva and intestinesand acidin
my stomachwould beginto transform the toast rapidly into glucose
shortly afteringestion.
Since my beta cells no longer produce detectable amounts of in
sulin, there is no storedinsulin to be released by my pancreas, so I have
no phase I insulin response. My blood sugar (in the absence of in
jected insulin) will rise while I digest my meal. None of the glucose
will be converted to fat, norwill anybe converted to glycogen. Eventu
allymuch will be filtered out by my kidneys and passed out through
the urine, but not beforemy body hasendureddamagingly highblood
sugar levels — which won't kill me on the spot but will do so over
a period of days if I don't inject insulin. The natural question is,
wouldn't injectedinsulin"cover" the carbohydrate in suchabreakfast?
Not adequately! This is acommon misconception— evenby those in
the health care professions. Injected insulin— even with an insulin
pump — doesn't work the same as insulin created naturally in the
body. Conventional insulin/diet therapyresulting in high blood sugar
after meals is a guaranteed slow, incremental,"silent" death from the
ravages of diabeticcomplications.
Normal phase I insulin is almost instantly in the bloodstream.
Rapidlyit beginsto hustleblood sugar off to whereit'sneeded. Injected
insulin, on the other hand, isinjectedeitherinto fat or muscle (not usu
allyinto avein) and absorbed slowly. The fastest insulin we have starts
to work in about 20 minutes, but its full effect is drawn out over a num
ber of hours, not nearly fast enough to prevent adamaging upswing in
blood sugars if fast-acting carbohydrate, like bread,is consumed.
This is the central problemfortype 1diabetics — the carbohydrate
andthe drastic surge it causes inbloodsugar. Because I knowmy body
produces essentially no insulin, I havea shot of insulin before every
meal. But I no longer eat mealswith fast-acting or large amounts of
carbohydrate, because the blood sugar swingsthey causedwere what
brought about my long-term complications. Eveninjection by means
of an insulin pump (seediscussion near the end of Chapter 19) can
not automaticallyfine-tune the levelof glucose in myblood the way a
nondiabetic's body does naturally.
Diabetes:TheBasics 47
Now, if I ate only the protein portion of the meal, my blood sugar
wouldn't have the huge, and potentiallytoxic, surge that carbohydrates
cause. It would riseless rapidly, and a small dose of insulin could act
quicklyenough tocover the glucose that's slowlyderived from the pro
tein. My body would not have toendure wide swings in blood sugar
levels. (Dietary fat, bythe way, has noeffect onblood sugar levels, ex
cept that it canslightlyslowthedigestion ofcarbohydrate.)
In a sense, youcould lookat myinsulin shot before eating onlythe
protein portion of the meal as mimicking the nondiabetic's phase II
response. This is much easier to accomplish than trying to mimic
phase I, because of the much lower levels of dietary carbohydrate
(only the slow-acting kind) andinjected insulin that I use.
The Type 2 Diabetic
Let's say Jim, a type 2 diabetic, is 6 feet tall and weighs 300 pounds,
much of which is centered around his midsection. Remember, at least
80 percent of type 2 diabetics are overweight. If Jim weighed only
170pounds, he might well be nondiabetic, but because he's insulin-
resistant, Jim's bodynolonger produces enough excess insulin to keep
his blood sugar levels normal.
The overweight tend to be insulin-resistant as a group, a condition
that's not only hereditary but alsodirectly relatedto the ratio of vis
ceraland total bodyfat to leanbodymass (muscle). Thehigherthisra
tio, the more insulin-resistant a person will be. Whether or not an
overweight individual is diabetic, his weight, intakeof carbohydrates,
andinsulinresistance alltendtomakehimproduceconsiderablymore
insulinthan a slenderpersonof similar ageand height (seeFigure 1-3,
page 48).
Many athletes, because of their lowfat mass and high percentage
of muscle, tend as a group to require and make lessinsulin than non-
athletes. An overweight type 2 diabetic like Jim, on the other hand,
typicallymakes two to three times as much insulin as the slender non
diabetic. In Jim's case, frommanyyears of having to overcompensate,
his pancreas has partially burned out, his ability to store insulinis di
minished or gone, and his phase I insulin responseis attenuated. De
spite his huge output of insulin, he no longer can keep his blood
sugars within normal ranges. (In my medical practice, a number of
patients come to me for treatment of their obesity, not diabetes. On
examination, however, most of these very obese"nondiabetics" have
slight elevations of theirHgbAlc test for average blood sugar.)
1 1
^^ObcscDonnal subjects
w^ Obe*patients wi h diabetes
' fcinn >rraal tubje » ""
V inpatient*' vithdiabcte i
0 15 30 45 60 90 120 150 18
Time (minutes)
Before You Start
Fig. 1-3. Serum insulin
response toglucose consump
tion of individuals withand
without type 2 diabetes.
Let's take another look at that mixed breakfast and see howit affects
atype 2 diabetic. Jimhasthe sametoastandjelly andjuiceandboiled
egg that Jane, our nondiabetic, and I had. Jim's blood sugar levels at
wakingmay be normal.* Since he hasabigger appetitethan either Jane
or I, he hastwo glasses of juice, four pieces of toast, andtwo eggs. As
soonasthetoastandjuice hit hismouth, hisbloodsugar beginsto rise.
Unlike mine, Jim's pancreas eventually releases insulin, but he hasvery
Httle or no stored insulin (his pancreas workshardjust to keep up his
basal insulin level), sohehas impaired phase I secretion. His phase II
insulinresponse, however, maybe partially intact. So, veryslowly, his
pancreas will struggle to produce enough insulin to bringhis blood
sugar down towardthe normal range. Eventually it may get there, but
not until hours after his meal, and hours after his body has been ex
posed to highbloodsugars. Insulin isnot onlythe major fat-building
hormone,it also serves to stimulate thecenters inthebrain responsible
for feeding behavior. Thus, in all likelihood, Jim will groweven more
overweight, asdemonstrated by the cycle illustrated in Figure 1-1.
Since he's resistant to insulin, his pancreas has to work that much
*Waking, or fasting, bloodsugars arefrequendy normalin mildtype2diabetics.
After theyeatcarbohydrate, however, their postprandial blood sugars areusually
Diabetes: The Basics 49
harder toproduce insulin toenable him toutilize the carbohydrate he
consumes. Because of insulin's fat-building properties, hisbodystores
away some of hisblood sugar asfat andglycogen; but hisblood sugar
continues to rise, since hiscells areunable to utilize all of the glucose
derived from his meal. Jim, therefore, still feels hungry. As he eats
more, hisbetacells work harder to produce more insulin. Theexcess
insulin and the"hungry" cells in his brain prompt him to want yet
more food. He has just one more piece oftoast with alittle more jelly
on it, hoping that it will be enough to get himthrough until lunch.
Meanwhile, his blood sugar goes even higher, his beta cells work
harder, and perhaps a few burn out.* Even after all this food, he still
may feel many of thesymptoms of hunger. His blood sugar, however,
will probablynot goanywhere nearashighasminewould if I tookno
insulin. In addition, hisphase II insulin response could even bringhis
bloodsugardownto normal aftermanyhourswithout more food.
Postprandial (after-eating) bloodsugarlevels that I wouldcall un-
acceptably high— 140 mg/dl, or even 200 mg/dl —maybe consid
ered by other physicians to be unworthy of treatment because the
patient still produces adequate insulin to bring them periodically
downto normal,or "acceptable,"ranges. IfJim, our type2diabetic, had
received intensive medical intervention before thebetacells of hispan
creas began to burn out, he would have slimmed down, brought his
bloodsugars intoline,andeased theburdenonhispancreas. Hemight
evenhave"cured" his diabetesbyslimmingdown, as I'veseenin several
patients. But many doctors might decide such "mildly" abnormal
blood sugars are only impaired glucosetolerance (IGT) and do little
morethan"watch"them.Again, it'smybeliefthat aggressive treatment
at an earlystagecan save most patientsconsiderable lost time and per
sonalagonybypreventing complications that willoccur if bloodsugar
levels are left unchecked. Such intervention can make subsequent
treatment ofwhat can remain— a milddisease — elegantlysimple.
I include some hopeful forecasts of future treatments in this first
chapterbecause asyou're learning howto controlyour diabetes, hope
*Betacellburnout canbe causedboth byoveractivityof the cells and bythe tox
icityof high glucose levels.
50 BeforeYou Start
is a valuable asset. But your hope should be realistic. Your best hope
for controlling your diabetes is normalizing your blood sugars now.
That does not mean that the future will not bring great things. Dia
betes research progresses on a daily basis, and I hope as much as you
do for a cure, but it's still on the horizon.
Researchers are currentlytrying to perfectmethods for replicating
insulin-producing pancreatic betacells inthe laboratory. Doingthis in
a fashion that'scomparatively easyandcost-effective should not be an
insurmountable task, and indeed the preliminary results arequite en
couraging. Once patients' cells are replicated, they canbe transplanted
backinto patientsto actually curetheirdiabetes. After suchtreatment,
unless you were to have another autoimmune event that would de
stroythese newbetacells, youwould, at leastin theory, remainnondi
abetic for the rest of your life. If you had another autoimmune attack,
you would simply have to receive more of your replicated cells. An
other very hopeful approach currendy undergoing clinical trials in
humans is the transformation of the precursors of beta cells(the cells
that line the ducts of the pancreas) into actualbeta cells without even
removingthem fromyour body. This may be achieved by the simple
intramuscular injection of a special protein and is now being tested
for efficacy and possible adverseeffects at three centers.
Another potential approach might be to insert the genes for insulin
production into liveror kidney cells. These arepotential opportunities
for a cure, and have successfullycured diabetes in rats, but there are
still obstacles to overcome.
Yet another approach to replacing lost beta cells has been used by
two competingcompanies to curediabetes in animals. The technique
involves a series of ordinaryinjections of proteinsthat stimulate the
remaining betacells to replicate until the lost oneshave beenreplaced.
A very promising newapproachrelieson the fact that most diabet
ics, even most type Is, have a few beta cells that still replicate. Their
immune systems, however, make white cells called killer T cells that
destroy the new beta cells as fast as they are made — or faster. If the
culprit T cells can be isolated, they can be replicatedand used to cre
ateantibodies that canbe injectedinto diabetics to destroyallof their
culprit T cells without impairing their overall immunity. A diabetic's
few remaining beta cells would then be able to replicate, eventually
curing the diabetes. It's possible that these new "former diabetics"
would require antibody injections every fewyears to prevent the ap
pearanceof more culprit T cells.
Diabetes: TheBasics 51
With respect to the replication of betacells, the catch for me and
other diabetics who nolonger have any insulin-producing capacity is
that the cells from which new beta cells would be repUcated ideally
should beyour own, and after more than six decades I mayhave none.
Had my diabetes been diagnosed, say, ayear earlier, orhadmy blood
sugars been immaculately controlled immediately upondiagnosis, the
injected insulin mighthave taken much of thestrain off myremaining
beta cells and allowed them to survive.
Many people (includingthe parents of diabetic children) view hav
ingto use insulin as a last straw, a final admission that they are (or
their child is) adiabetic and seriously ill. Therefore theywill try any
thingelse — including things thatwill burnout their remaining beta
cells — beforeusing insulin. Many peoplein our culturehavethe no
tion that you cannot be well if you are usingmedication. This is non
sense, but some patients are soconvinced that they must do thingsthe
"natural" waythat I practically have to begthem to useinsulin, which
is as "natural" asone cango. In reality, nothingcouldbe more natural.
Diabetics who still have beta cell function left may well be carrying
their own cure around with them — providedthey don't burn it out
with high blood sugars and the refusal to use insulin.
Thegoal in the treatment program laid out in this book is to
give you the tools and the knowledge to take control of your
disease by normalizingblood sugars. Myinterest is not just in
treating the symptoms of diabetes, but in preventing or reversingits
consequences and preservingpancreatic beta cell function. Essential
to treatment is learning to monitor your own blood sugars.
Before youbeginto monitor and then normalizeyour blood sugars,
you should have a baseline analysis of your disease. How much have
your beta cells "burned out" in part fromhighblood sugars? Haveyou
already developed some easilymeasured long-term compUcations of
diabetes?What are your risksfor other diabetic compUcations?
Answering thesequestions willaidyouand your doctor in learning
the extent and the consequences of the disease. Your test results wiU
also serve as valuable baseline data to which you will be able to com
pare the effects of bloodsugarnormalization. Onceyour bloodsugars
havebeen normalized,suchtestscanbe repeatedfromtime to time, to
show what you're achieving. Your improvements wiU give both you
and your physicianongoingincentive for stickingto the program.
The remainder of this chapterdescribes a number of testsyour doc
tor or hislaboratorycanperformin order to give both of youa picture
of your diabetic condition. I havelaid these out notbecause it's neces
sary for you to memorize them, research them, and know aU the ins
and outs of them, but so that you can get the treatment you deserve.
Byoutlining these tests,I'm giving you a "shoppinglist"of tests I per
form on myself and on my patients.
Generally, I recommend as many as you can afford or your insur
ance or health maintenance organization (HMO) will pay for. Com-
Tests: Baseline Measures 53
pleting more of thetests wiU addmore dimensions to the picture you
gain ofyourdisease. As some of these tests arecostly, anyor allmaybe
skipped if youcannot afford themor if your insurance or HMOwon't
pay for them.
It is your physician's obUgation to provide you with copies of aU
your test results, whether fromlaboratory tests or from physical ex
aminations. This is your right; however, you must request them. The
laws governing medical records varystatebystate, and legislatures are
Ustening to health care consumers and making changes regularly. At
this writing, however, it is most often the casethat medical records are
the propertyof the doctor, so do not neglect to request copies of any
results. Such results can be potentially of great value when you visit
another physician or speciaUst for treatment of anyproblems.
Glycated Hemoglobin (HgbA,c)
Glucose binds to hemoglobin (the pigment of red blood ceUs) when
new red ceUs are manufactured. Since the average red ceU survives
about four months, the percentage of hemoglobinmoleculesthat con
tain glucose (HgbAlc) provides an estimate of average blood sugar
over this time frame. One of the benefits ofthis test is that it givesyour
physician an index by which to test the accuracy of your own blood
glucoseself-monitoring results. If your measurements are stricdy nor
mal but your HgbAlc is elevated, then your doctor has a clue that
something is awry.
There are, however, a couple of significant drawbacks to this test.
First is that the test is onlya measureof average blood sugars. Second,
elevatedblood sugars may take 24 hours to have any long-term effect
onHgbAlc, andifblood sugar iselevated for onlypartofeach day and
isnormalized or too lowthe restof thetime,your HgbA,c results may
appear deceptivelylow. Thus, if your blood sugars are only elevated
for a few hours after meals, your HgbAlc may not be affected, but
many tissuesand organs throughout your bodywill be injured.
The other drawback is that the upper and lower ranges of"normal"
values reported by most labs are usuallyerroneously high and low, re
spectively. In other words, the ranges are usuallymuch too wide. Thus,
it's up to your physicianto decide, basedupon his experience, what the
proper normal range for his lab should be. Some doctors have their
54 BeforeYou Start
own formulas for estimating average four-month blood sugar levels
from HgbAlc. Anormal value should correspond to blood sugars of
about 80-95 mg/dl. The experience I've had with the lab I use (the
largest in the United States) for my patients is that a truly normal
HgbAlc ranges from 4.2 percent to 4.6 percent, which corresponds to
blood sugarsof about 83-90 mg/dl. Mine is consistently4.5percent. A
recent study of "nondiabetics"showeda 28 percent increase in mor-
taUty for every 1percent increase in HgbAlc above 4.9 percent.
Because the blood contains more recentlymade red cellsthan older
ones, recent blood sugars have more of an effect on HgbAlc thando
earUer blood sugars. The test value therefore levels off after about
three months. Anyailment that hastens red blood ceUloss will cause a
deceptive shortening of thetime frame reflected bytheHgbAlc. Such
ailments include Uver and kidney disease, blood loss, hemoglo
binopathies, et cetera. High dosesof vitamins Cand Ecan cause a de
Serum C-Peptide (Fasting)
C-peptide is a protein produced by the beta cells of the pancreas
whenever insulin is made. The level of C-peptide in the blood is a
crude index of the amount of insulin you're producing. The level is
usuaUy zero in type 1 diabetics, and within or above the "normal
range" in mildtype2obese (insulin-resistant) diabetics. If your blood
serum C-peptide is elevated, this wouldsuggest to your physician that
your blood sugar maybe controllable merely bydiet, weightloss,and
exercise. If, at the other extreme, your C-peptide is belowthe limits of
measurabiUty, you probably require injected insulin for bloodsugar
normalization. C-peptidemeasurements, to bemost significant, should
be checked aftera 12-hourfast whenbloodsugars arenormal. Thetest
can be best interpreted if blood sugar is measured at the same time,
because in nondiabetics high blood sugars cause more insulin (and
C-peptide) productionthan do lowbloodsugars.
Thistest, while of interest, is not absolutely necessary.
Complete Blood Count (CBC)
Part of most medical workups, thisisa routinediagnostic test that can
disclose the presence of ailments other than diabetes. ACBCmeasures
the number of various typesof ceUs found in your blood — white cells,
red ceUs, and platelets. Ahigh level of white bloodcells, for example,
can disclose the presence of infection, while too few red blood cells
Tests: BaselineMeasures 55
can indicate iron deficiency anemia. Many diabetics have inherited
thyroid dysfunction, which can cause low-normal to low white ceU
counts. Awhite ceU countless than5.6 suggests that afull thyroid pro
file should be performed. This must include free and total T3 and T4.*
ACBCcanalsodetectcertainhematologic maUgnancies, whichare
usually moreeffectively treated theearner they arediscovered.
Standard Blood Chemistry Profile
This batteryof twelve to twenty tests is part of most routine medical
examinations. It includes gaugesfor such important chemical indica
tors of healthasUver enzymes, bloodureanitrogen(BUN), creatinine,
alkaline phosphatase, calcium, andothers. Ifyouhave a history of hy
pertension, your doctor maywant to add redbloodceUmagnesiumto
this profile.
Serum Albumin
Although serum albumin is usuaUy included in the blood chemistry
profile, it is not widely appreciated that lowlevels are associated with
double the all-cause mortaUty of high levels. It is thus veryimportant
that patients with lowserum albumin receive further workup to de
termine the cause.
Serum Globulin
Globulins are antibodies producedbythe immune system. Theyhelp
the body to fight off infections and maUgnancy. If you experience fre
quent colds, sinusitis, diarrhea, or slow-healing infection of any type,
you may have an immunoglobulin deficiency. If your total serum
globulinsare low, you should be testedfor specific immunoglobulins,
such as IgA, IgG,and IgM.In myexperience, 10-20 percent of diabet
ics have an inherited immunodeficiency disorder that may be treat
Cardiac Risk Factors
This is a battery of teststhat measuresubstances in the blood that may
predispose you to arterial and heart disease.
* It's worth noting that one of the hallmarks of high blood sugars is fatigue.
However, diminished thyroid function can causeprofound fatigue and coldness.
If you're still "always tired" or "always cold" after normalizing blood sugars, talk
to your physician about a thyroid profile. This test can be costly.
56 Before You Start
Important note: Sometimes, months to years after a patient has
experienced normal or near-normal blood sugars and resultant im
provements inthecardiac riskprofile, we might see deterioration inthe
resultsof such testsas those for LDL, HDL,homocysteine, fibrinogen,
andUpoprotein(a). All toooften, thepatient orhis physicianwill blame
our diet.Inevitably, however, wefindupon further testing that his thy
roidactivityhasdeclined. Hypothyroidismisanautoimmune disorder,
like type 1 diabetes, and is frequently inherited by diabetics and their
close relatives. It can appear years beforeor after the development of
diabetes and is not caused by high blood sugars. In fact, hypothy
roidismcan cause a greater likelihood of abnormalities of the cardiac
risk profile than can blood sugar elevation. The treatment of a low
thyroid condition is oral replacement of the deficient hormone(s) —
usually 1 piU daily. The best screening test is free T3 as measured by
tracer dialysis. If this is low, then a full thyroid test profile should be
performed. Correctionof the thyroiddeficiency inevitably correctsthe
abnormaUties of cardiac risk factors that it caused.
Lipid profile. This profile measures fattysubstances(lipids) in your
blood and includes total cholesterol, HDL(high-density lipoprotein),
triglycerides, and direct LDL (low-density Upoprotein). Other cardiac
risk factors (discussedbelow) include C-reactive protein, fibrinogen,
lipoprotein(a), and homocysteine, and may be more predictive. Ab
normalities of these tests are frequently treatable and tend to improve
with normalization of blood sugars.
These tests should be performed after you have fasted for at least
8 hours. The easiest thing is to have them scheduled in the morning.
If you haven't fastedbefore the test, the results will be difficult to in
Maybeyou've heard of"good"cholesteroland "bad" cholesterol?
Well,this is why a reading for total cholesterol by itself won't neces
sarily reflect cardiac risk. Most of the cholesterol in our bodies, both
good and bad, is made in the liver; it does not come from eating so-
caUed"heart attackfoods." If you'veeatena meal that's high in choles
terol, your Uver will adjust to makelessof the "bad" cholesterol, LDL.
Serum triglyceridelevels can vary dramaticallyafter meals, with high-
carbohydrate meals causing high triglyceride levels. Some people —
becausethey're obeseor havehighblood sugarsor are genetically pre
disposed— make more or dispose of less LDL than they should,
Tests:Baseline Measures 57
which can put them at a higher risk for cardiac problems. High levels
of LDL increase the risk of heart disease, which makes LDLthe "bad"
cholesterol. HDL, on theotherhand, isa Upid that reduces the riskof
heart disease and is the "good"cholesterol. So it is the ratio of total
cholesterol to HDL (total cholesterol -s- HDL) that is significant. You
couldhave a high total cholesterol and yet, because of lowLDL and
high HDL, have a lowcardiac risk. Conversely, a low total cholesterol
butwith alowHDLwould signifyincreased risk. Recently, asmorehas
become known about cholesterol, research has shown that LDLoccurs
in at least two forms—smaU, dense LDL particles (the hazardous
form) andlarge, buoyant LDL particles. LDL particle size isnowbeing
measured by commercial laboratories. Larger particles, classified as
size A, are considered benign, while smaller particles carry cardiac
risk. Associated with the test for particle size is apoUpoprotein B.
When theApo Btestresult islower than120 mg/dl, or when LDL par
ticle size is class A, even high LDL levels are considered benign and
shouldnot be treatedwithstatindrugs.
The only truly accurate measureof LDL is the direct LDL test. The
customary, calculated measure of LDL is estimated mathematicaUy
and can result in values that aresometimes grossly in error. The direct
test,however, maycost more than the rest of your Upid profile.
Also important to remember is that —aswewiU discuss in Chap
ter 9— fats and cholesterol in the diet do not cause high-risk Upid
profiles in most people. On theotherhand,diabetics tendto have Upid
profiles that reflect increased cardiac risk, if their blood sugars have
been elevated for several weeks or months.
Thrombotic risk profile. Thisprofile includes levels of fibrinogen,
C-reactive protein, and Upoprotein(a). Theseare also"acute phase re-
actants," or substances that reflect ongoing infection and other in
flammation. These three substances are associated with increased
tendency of blood to clot or form infarcts (blockages of arteries) in
peoplewho havehad sustainedhigh blood sugars.
In the casesof elevatedfibrinogenor Upoprotein(a), there is, addi-
tionaUy, often an increased risk of kidney impairment or retinal dis
ease. Obesity, even without diabetes, can cause elevation of C-reactive
protein. In my experience, aU these tests are more potent indicators of
impending heart attackthan the Upid profile. Treatmentsare available
for elevationsof each of these. Bloodsugar normalization wiU tend to
58 BeforeYou Start
reverse most of theseelevations overthe longterm. Fibrinogencan be
elevated bykidney disease, even in the absence of elevated bloodsug
ars. It wiU tend to normalizeif kidneydisease reverses. Lipoprotein(a)
wiU also tend to normalize somewhat by blood sugar normalization,
although your genetic makeup (and lowestrogen levels in women)
canplaya greater role thanbloodsugar. AbnormaUy lowthyroidfunc
tion is a common cause of lowHDLand elevatedLDL,homocysteine,
and Upoprotein(a). Although serum homocysteine is also a cardiac
riskfactor, it wasrecently discovered that the usual treatment for ele
vated values (vitamin B-12 and foUc acid supplements) actuaUyin
creased mortaUty.
Serum transferrin saturation, ferritin, total iron binding ca
pacity (TIBC). These are aU measures of total body iron stores. Iron
isvital,but it isalsopotentiallydangerous. Levels that are too high can
indicate a cardiac risk, cause insulin resistance, and are a risk factor for
Uver cancer. I wiU discuss insulin resistance at length in Chapter 6.
Higher iron levels are more likely in men than in premenopausal
women because of blood (iron) loss during menstruation. (This is
why I recommend iron-enhanced vitamin supplements onlyfor those
with an established need.) Iron levelsthat are too low (iron deficiency
anemia, which is more common in premenopausal women) can cause
an uncontroUableurge to snack, which in turn can lead to uncontrol
lable blood sugars. Both high and lowiron stores can be easily deter
mined and readily treated.
Renal Risk Profile
Chronic blood sugar elevationfor many yearscan cause slowdeterio
ration of the kidneys. If caught early, it may be reversible by blood
sugar normalization, as it was in my own case. Unless you think fre
quent hospital visits for dialysis might be a nice wayto meet people,
it's wise to have periodic tests that reflect early kidney changes. It is
also wise to have aU these periodicaUy performed together, as the re
sults of each can clarifythe interpretation of aU.
Several factors cause false positiveresults in some of these tests, so
you should keep them in mind when your doctor schedulesthe tests.
Youshould avoid strenuous or prolonged lower-body exercise (which
would include motorcycle or horseback riding) in the 48 hours pre
ceding the tests. AdditionaUy, if on the day the tests are to be per
formed you are menstruating or havea fever, a urinary tract infection,
Tests:Baseline Measures 59
or active kidneystones, youshouldpostponethe testsuntil thesecon
ditions have cleared.
Abasic renal riskprofile should include the foUowing:
Urinary kappa light chains. Ifearlydiabetic kidneydisease ispres
ent, this test reports "polyclonal kappa Ught chains present." This
means that small amounts of tinyprotein molecules maybe entering
the urine,duetoleakybloodvessels inthekidneys. Because thesemol
ecules aresosmaU, theyarethefirst proteins toleak throughtinypores
in the bloodvessels of the kidneys that mayhave beenaffected bydis
This test requires a smaU amount of fresh urine. If the test report
states "monoclonal Ught chains present," thereis a possibiUty of treat
ablemaUgnancies of certain whiteblood ceUs.
Microalbuminuria. Thisless cosdytest cannowbe performedqual
itatively (bydipstick) in your doctor's office, or quantitatively at out
side laboratories. It, like the urinary kappa Ught chain test, can also
reflect leakyvessels in the kidneys, but at a laterstage, sincealbuminis
a sUghtly larger molecule.
A quantitative measurement requires a 24-hour urine specimen,
whichmeans you'U need to coUect aU the urine you produce in a 24-
hour period in a big jug and deUver it to your physicianor laboratory.
Given the potential embarrassment of carrying a jug fuU of urine
around at work, you might want to schedule your test on a Monday
and coUect the urine while at home on Sunday. Many of my women
patients report that it's easierto coUect urine initiaUy in a clean paper
cup, and then pour it into the jug. An easier screening test is the mea
surement of the albumin-to-creatinine ratio in a first morning urine
24-hour urinary protein. This test detectskidneydamage at a later
stagethan the preceding two tests; it alsorequires a 24-hour urine col
lection.Aswith the other tests, false positive resultscan occur foUow
ing strenuous lower-body exercise, as previouslynoted.
Creatinine clearance. Creatinine is a chemical by-product of mus
cle metabolism, and is present in your bloodstream aU the time. Mea
suring the clearanceofcreatinine fromthe body is a wayof estimating
the filteringcapacityof the kidneys. Test values are usuaUy higher than
60 BeforeYou Start
normal when a person is spilling a lot of sugar in the urine, and even-
tuaUy lower than normal when the kidneys have been damaged by
years of elevated bloodsugars. It is not surprising to seean appropri
ate drop in creatinine clearance when blood sugars are normalized
and urine glucose vanishes.
The creatinine clearance test requires a 24-hour urine coUection,
andyour doctor wiU drawa smaUamount of bloodto measure serum
creatinine. The most common causeof abnormaUy lowvaluesfor this
test is failure of the patient to coUect aU the urine produced in a
24-hour period. Therefore, if other kidneytests are normal, testswith
lowvalues for creatinine clearanceshould be repeated for verification.
Alow creatinine clearancewithout excess urinary protein suggests
a nondiabetic cause of kidney impairment.
When it is impractical to makea 24-hour urine collection, as for small
children, a new test requiring a smaU amount of blood, crystatin-c,
can be performed. Crystatin-c is beUeved to be a more accurate mea
sure of kidney function than creatinine clearance, but unfortunately
many insurers still consider this test to be "experimental" and won't
pay for it.
Serum beta2 microglobulin. This is a verysensitive test for injury
to the tubules ofthe kidneys, which pass urine filteredfrom the blood.
As with fibrinogen levels, elevatedvaluescan also result from inflam
mation or infection anywhere in the body. Thus an isolated elevation
of serum beta2 microglobulin without the presence of urinary kappa
Ught chains or microalbumin is probably due to some sort of infec
tion, not to diabetic kidneydisease. Suchelevationis commonplace in
people with AIDS, lymphoma, and immunodeficiencydisorders.
24-hour urinary glucose. This test too requires a 24-hour coUection
of urine, and is of value for proper interpretation of the creatinine
Note: If, as you've been reading about these tests, you've imagined
yourself lugging around multiple jugs of urine, most of us only need
one 3-liter jug. This should give you an adequate specimen for your
physician to perform creatinine clearance, microalbumin, 24-hour
protein, and 24-hour glucose. Nevertheless, it's wise to bring home
two empty jugs, just in caseyour urine output is very high.
Tests: Baseline Measures 61
As indicated under "Cardiac Risk Factors," significant kidney dam
age isalso accompanied byelevations of serum homocysteine and fi
Insulin-like Growth Factor 1 (IGF-1)
Rapid correction of veryhighbloodsugars can,on occasion, cause ex
acerbation of a common compUcation of diabetes caUed proliferative
retinopathy. This condition cancause hemorrhaging inside theeye and
blindness. Such exacerbations are usuaUy preceded by an increase in
serumlevels of insulin-like growth factor 1.Abaseline level of IGF-1 in
theblood should bemeasured inpeople with proliferative retinopathy.
Repeat determinations shouldbe madeevery twoto three months. If
levels increase, bloodsugars should thenbereduced moreslowly.
R-R Interval Study
Thepurposeof this studyis to test the functioning of the vagus nerve,
and it should be part of your initial diabeticphysical examination. It is
performed like an ordinary electrocardiogram, but it requires fewer
electrical leads (i.e.,onlyon the limbs, not the chest).
The vagus nerveis the largest nerve in the body, running fromthe
brain to the lower body. It's the main neural component of the
parasympathetic nervous system, or that part of the nervous system
that takes care of vegetative, or autonomic, functions, the functions
that run more or less on "automatic pilot" and which you don't ac
tively have to think about to make happen. These include heart rate
and digestion.
Like anyother nerve in the body, the vagus nervecan be injured by
long-termexposure to highbloodsugars, but since it plays sucha cen
tral role in bodily function, damage to it can cause many more disor
ders than damage to most other individual nerves.
The vagusplaysa major rolein a number of diabeticcompUcations
involving the autonomic nervous system, including rapid heart rate,
erectile dysfunctionin men, and digestive problems, particularlygas-
troparesis, or delayedstomach-emptying (which we wiU discuss in de
tail in Chapter 22). The good newsis that when you'vehad your blood
sugars normalized over an extended period, it can slowly recover
62 Before You Start
proper function. (Many ofmy male patients who have been unable to
achieve or maintain an erection report that after blood sugars have
been normalized,that abiUty has returned.)
This nerve is unique in that itsfunction canbe investigated simply
andcheaply. If thevagus nerve isworking properly, there should bea
considerable difference in heart rate between inhaling and exhaling.
By measuring thevariation ofyour heart rate with deep breathing, we
canget a picture ofjusthowmuch thefunction has been impaired. In
nondiabetics, the heart rate increases when they inhale deeply and
slows when theyexhale fully. So a twenty-one-year-old nondiabetic's
heart ratemight typicaUy slowasmuch as75 percent from inhaling to
exhaling. This may drop to about 30 percent for a seventy-year-old
nondiabetic. Ayoung type 1diabetic withtenyears ofveryhighblood
sugars maynot have anyheart ratevariation at aU. (Thevariation is
measuredby lookingat the interval between "R-points," or peaks on
the tracing of the electrocardiogram. As you'reprobablyaware, each
time your heart beats, the electrocardiograph traces a shape resem
blinga mountain.Thetipof themountainistheR-point, sothe physi
cian measures the intervalsbetweenR-points.)
I consider this test an important, reproducible, quantitative mea
sure of an important diabetic compUcation and perform it on aU of
my newpatients beforeblood sugars have been stabUized. I repeat it
about everyeighteenmonths, for several reasons. It's a verygood in
dexof how,with aggressive blood sugar control, neurologiccompUca
tions can and do reverse, and it gives both patient and doctor good,
concrete evidence of the success of treatment, and encouragement to
keepit up. Additionally, the digestive disorder of gastroparesis, which
I mentioned above, can be and frequently is one of the most difficult
barriersto bloodsugarnormalization, andcanevenmakebloodsugar
control virtuaUyimpossiblein some peoplewho require insulin. Alow
heart rate variabiUtyon the initial test can be a good indicator that the
patient is likely to have a problemwith delayed stomach-emptying. It
can alsogive the doctor clues asto causes of other problemsthat a pa
tient maybe experiencing —sexual dysfunction, faintingupon stand
ing when arising from bed, and so on.
Neurologic Examination
In addition to a standard physical examination,it is desirable(but not
essential) that a routine neurologic exam be performed before blood
sugars are corrected, and againeveryfew years thereafter. These tests
Tests: BaselineMeasures 63
arenot painful. Theyshould include checking forsensation inthefeet,
reflexes oflimbs and eyes, short-term memory, and muscle strength.
In myexperience, performance on a number of the neurologic tests
improves after many months of essentiaUy normal blood sugars. Per
formance tends todeteriorate ifblood sugars remain high.
Eye Examination
One of the most valuable retinal studies, the Amsler grid test, can be
performedbyanyphysician or nursein less than a minute without di
lating your pupUs. Since chronicaUy high blood sugars frequently
cause a number of disorders that canimpairvision, your eyes, if nor
mal, shouldbeexamined carefuUybyanophthalmologist every oneto
two years.
The ophthalmologist wiU evaluate the retina, lens, and anterior
chamber in each eye, and you can expect to have your pupils dilated
with special drops. Aproper retinalexamrequires the useof both di
rect and indirect ophthalmoscopes and a sUt lamp.If an abnormaUty
isfound, certainexaminations mayhave to be performedbya retinol-
ogist everyfewmonths.
Examination of the Feet
Because ulcers of the diabetic foot are avoidable, even when blood
sugar is not weU controUed, you should askyour physicianto examine
your feet at every routine officevisit. Foot problems that aren't pre
vented or treated properly can leadto serious compUcations, evenam
putation. Your physician should train you in foot self-examination
and preventive care. In Appendix D, I have reproduced the same in
structions I give my own patients on how to care for their feet.
Oscillometric Study of Lower Extremities
This inexpensive test utilizes a simple blood pressure cuff connected
to a small instrument that should be in everydoctor's office.It givesan
indexof the adequacyofpulsatilecirculationto the legsand feet. Since
long-standing, poorly controUed diabetes can seriously impair pe
ripheral circulation, this test is fairlyimportant. AU diabetics should
take special care of their feet, but if you have an abnormal osciUomet-
ric study, you have to be extra careful. People who have diminished
circulation in the legs usuaUy also have significant deposits in the ar
teries that nourish the heart and brain and the arteries necessary for
penile erection. Therefore, if this study shows impaired circulation,
64 Before You Start
your doctor may want you to undergo tests that would help diagnose
coronary artery disease and, if you have certain symptoms, diagnose
impaired circulation to the brain. OsciUometry canbe performed by
anytrainedphysician inbut a few minutes. It istaughtat manymed
ical schools throughout the world but rarely in the United States,
where hands-on care is diminishing.
Musculoskeletal Examination
Prolonged high blood sugars can cause glycation of tendons. Glyca
tion is the permanent fusing of glucose to proteins, and the simplest
analogy is bread crust. Think of the soft inside of the bread as your
tendons astheyshouldbe, and the crust aswhat happens whenthey're
exposedto elevated bloodsugars overa longperiod of time. Glycation
of tendons occurs in such common diabetic complications as Du-
puytren's contractures of the fingers, frozen shoulders, triggerfingers,
carpal tunnel syndrome, and iliotibial band/tensor fascialata syn
drome of the hips and upper legs. All of these conditions are easily
treated if caught early and blood sugars are controUed. A muscu
loskeletal examination can identify these in their early, treatable
When to Perform These Tests
As valuable as they can be to you and your physician, none of the
abovetests is crucial to our central goal of achieving blood sugar nor
malization. If you are without medicalinsurance, or if your insurance
won't pay for these tests, and financial considerations are a top prior
ity, aU can be deferred. If, however, you are experiencing problems,
such as impairment of vision, youshouldbe testedimmediately. Also,
examinationof your feet, and learninghowto carefor them properly,
is vital and can prevent or forestall serious problems.
The most valuable of these tests for our purposes is the HgbAlc,
because it alerts your physician to the possibility that your self-
monitoredbloodsugardatamaynot reflect the average bloodsugarfor
the prior three months. This can occur if your blood sugar-measuring
technique or supplies are defective. More commonly, some patients,
with a scheduled visit to the doctor approaching, wiU improve their
* For information on the proper treatment of these conditions, visit www.
diabetes-book.com; select"Articles" and then the article tide that begins "Some
Long-TermSequelaeof PoorlyControlled Diabetes."
Tests: Baseline Measures 65
eating habits so that their blood sugar records improve. I have seen
several teenagers whose falsified blood sugar data were discovered by
this test. Itherefore suggest that HgbAlc be measured at regular visits
every two to three months. This test costs about $65.
IdeaUy, the other blood and urine tests should beperformed before
attempting to normalize blood sugar and annuaUy thereafter. If anab
normal value is found, your physician maywish torepeat thattestand
related tests more often. The exception isthe fasting C-peptide test, as
there islittle value inrepeating it except tosee if pancreatic function is
deteriorating or improving. I certainly like to repeat the thrombotic
risk and Upid profiles about four months and theneight monthsafter
blood sugars have been normalized. The improvement that I fre
quently see tends to encourage patients to continue their efforts at
blood sugar normalization. The R-Rinterval test shouldbeperformed
every eighteen months. I consider it the second most important test I
perform on my patients after the HgbAlc
A final note: Dietaryvitamin Cisimportant to goodhealth.In doses
above 500 mg/day, however, vitamin C supplements can destroy the
enzymeson bloodsugar test strips andcan also raise bloodsugars. Fi-
naUy, in levels higherthan about 400 mg/day, vitamin C becomes an
oxidantrather thananantioxidant andcan cause neuropathies. If you
are already takingsupplemental vitaminC, I urgeyouto taperit off or
lower your dose to no more than 250 mg daily. Only use the timed-
release form.
Your Diabetic Tool Kit
Inorder to monitor and control your blood sugarlevels,you're go
ing to need certain tools. This chapter lists and describes them;
you'U learnmore about them in laterchapters. Also included are
suppUes for foot care and for treating dehydrating illnesses. For most
items, approximate costs are Usted. Some expenses wiU be onetime
outlays, such as for your blood glucose meter outfit. Others wiU con
tinue on an ongoing basis.
The tools that all diabetics wiUneed areUsted first. Tools that only
insuUn users wiUneed are Usted separately. Some are necessary, and
some areoptional. You can show your physicianthe list and he or she
candecidewhich items are appropriate for your needs.
FoUowing the tablesof supplies is abrief description of each, what
it's for, where you can purchase it, whether you'U need a prescription,
and where in this book you'U find a complete description of its use.
If you can't locate some of these suppUes in your area, aU pres
cription items and most nonprescription items can be ordered via
telephone and credit card,check, or money order from RosedalePhar
macy, (888) 796-3348.
Your Diabetic Tool Kit
Formeasuring and recording
blood sugar
Blood sugar meteroutfit (including
lancingdeviceand lancets)
Blood sugartest strips (at least 1boxof 50)
Glucograf III data sheets
Forremoving blood from clothing
For dehydration
Morton's LiteSalt, Featherweight Salt
Substitute, Diamel Salt-It, Adolf's Salt
Substitute, Nu-Salt Salt Substitute, etc.
For diarrhea
Lomotil (diphenoxylate HCl with
atropine sulfite)
Approximate cost
$12.95/pad of 26double-sided
sheets(covers one year),at
Rosedale Pharmacy
$25/60ml dropper bottle; $4/100
tablets (genericprices)
For severe vomiting
Tigansuppositories (trimethobenzamide HCl) $45/box of 10,generic, 100mg
(children), 200 mg (adults)
For low blood sugars (required if taking medication
that lowers blood sugar)
Dextrotabs: 1-3 bottles of 100 tablets. $13/bottle of Dextrotabs
Also,in an emergency, B-Dglucosetablets,
Dextro Energy,Dextro-Energen, Dex4, Sweetarts,
or Winkies (kosher)
MedicAlert identificationbracelet $35-$700(gold)
For urine testing during Illness
Ketostix (foil-wrapped), 1package $9.50
*Seeyour pharmacist for current mail-in rebate or trade-in deals.
f This is an important product. The bracelet and accompanying contact infor
mation (see page 71) can provide paramedics or other medical professionals
with considerable information in case of loss of consciousness.
Before You Start
For testing food for sugar
Clinistixor Diastix, 1package
For foot care
Olive oil, vitamin Ecream, coconut oil,
mink oil, emu oil, etc.
Bath thermometer
For menu planning
The Complete Book ofFood Counts, 7thed.,
$7.99 (paperback)
Corinne T. Netzer (Dell, 2005)
Bowes &Church's Food Values ofPortions
$59.95(plasticcomb bound)
Commonly Used, 17thed., JeanA.T.Penning
ton, ed. (Lippincott Williams &Wilkins, 2004)
The NutriBase Complete Book ofFood Counts $14.95 (paperback)
(Avery, 2001)
Artificial sweeteners
Stevia extract $18.95/2ounces liquid; $18/
1ounce powder
Saccharin tablets $4.80/1,000 tablets (1/2 grain)
Equal tablets $2.95/100 tablets
Insulins and insulin supplies*
Humalog (Lilly) or Novolog(Novo), 2 vials $84/vial
Humulin R (Lilly) or Novolin R (Novo), 2 vials $39/vial
Lantus insulin, 2 vials $90/vial
Levemir (Novo), 2 vials $90/vial
Frio (to store insulin whiletravelingin $29
hot climates)
30-unit short-needle insulin syringeswith $29/box of 100
Vi-unit markings, such as B-DUltrafineII
(get at least 200to start)
30-unit standard-needle insulin syringessuch $29/boxofl00
as B-D Ultrafine (for correcting elevated
blood sugarswith optional intramuscular
injections— get 100to start)
*The particular insulins to be used willvary from one person to another, as in
dicated in Chapters 17-19.
YourDiabetic Tool Kit 69
For low blood sugar emergencies
Glutose 15 (Paddock Laboratories) $15/3tubes
Glutolliquidglucose* (PaddockLaboratories) $6/bottle
GlucagonEmergencyKit $104
Metoclopramidesyrup $20/4-ouncebottle
For All Diabetics
Blood sugar meter outfit. Ablood sugar meter outfit should con
tain abloodsugarmeter, a finger-stick device, and a smallstartup sup
plyof disposable lancets and test strips. The meter does the samejob
as the one I bought decades ago, althoughmyold one weighed three
pounds and some of the newmodels have footprints smaller than a
credit card. They work quite simply: with one drop of blood from a
finger stick, the instrument gives you a reading of your blood sugar.
(See Chapter 4, "How and When to MeasureBloodSugar.")
For small children and for people who develop calluseson their fin
gertips from frequent blood sugar testing, try Vaculance (Bayer),
whichcan securea drop of blood fromarms, buttocks, or abdomen. I
use it myself occasionally.
Blood sugar meters are available at most pharmacies and chain
drugstores. Some are more accurate and reliable than others. Because
of the rapid advances in technology, it wouldbe counterproductive to
recommend a particular meter in this book.* If you want our current
recommendation, please call our Diabetes Center at (914) 698-7525,
Monday through Thursday, 9:30a.m. to 3:00p.m. Eastern (U.S.) time.
Disposable 30-gauge lancets. Theseare usedwith or without your
finger-stick device to puncture your skin for glucose testing. I reuse
mine until they become dull. The small supply packed with the vari
ous meter outfits should easilylast a year.
* For people with severegastroparesis(seeChapter 22).
+Themost accuratepersonalmeter asof thiswritingis the HemoCue. It is much
larger and more difficult to use than other meters and requires much more
blood. It is great for calibrating more portable meters.
70 BeforeYou Start
Blood sugar test strips. When youstick yourfinger, you'll put the
dropof blood onor into oneof these. They work with your meter to
give yourblood sugar readings.
Glucograf HI data sheets. SeeChapter 5, "RecordingBlood Sugar
Data."These are essential to record your blood sugars and other im
portant data properly. They are available from Rosedale Pharmacy
and www.rx4betterhealth.com. These are the only data sheets that I
Hydrogen peroxide. Now andthen, whenyou're sticking your fin
gers or injecting through your shirt, asI do, you may get a little blood
onyourclothing. Hydrogen peroxide isaneffective waytoeradicate it.
You can get small bottles for home, office, car, or travel. Available at
anydrugstore andat many groceries. (See Chapter 16, page 261.)
For dehydration. Dehydrating illnesses, suchasvomiting,diarrhea,
or fever, are potentially fatal for diabetics. If youbecome dehydrated,
theseproducts canhelpyou replace lost electrolytes. Look for potas
sium chloride on the list of ingredients. These should be available at
the supermarket or grocery store. Their useis covered in Chapter21,
"Howto Copewith Dehydration, Dehydrating Illness, and Infection."
Theyshouldbe used asdirected bya physician.
For diarrhea. Diarrheacancausedehydration.The one product that
appears always to workis Lomotil (diphenoxylate HCl with atropine
sulfate). This is a prescription drug. Generic versions are available at
lower cost.
For vomiting. Vomiting can also cause dehydration, which, as noted
above, can be life-threatening for diabetics. Tigan suppositories (tri
methobenzamide HCl) frequently relieve vomiting and should be
usedfor no more than 1-2 days at a time, unless directedotherwiseby
your physician. Theyare available in 100 mg (children and elderly)
and 200 mg (adults) dosages. Use rectally as directed in Chapter 21.
This product requires a prescription. Generic versions are available.
For low blood sugars. If you experience low blood sugars, Dex
trotabs are a good, controlledway of bringingthem up by precisein
crements, while minimizing the risk of overshoot that you might
experience with, say, a glass of fruit juiceor a soft drink. Each tablet
YourDiabetic Tool Kit 71
will raise theblood sugar of a 140-pound person byabout8 mg/dl. If
yourun out of Dextrotabs and need an emergency supply of glucose
tablets, youcan use one of the following: B-D glucose tablets, which
will bring up blood sugars approximately 25 mg/dl in most adults;
Dextro Energy, approximately 15 mg/dl; Dextro-Energen, 20 mg/dl;
Dex 4,20 mg/dl; Sweetarts, 10 mg/dl;* andWinkies, 2mg/dl. Most of
these are available through most drugstores; Winkies are onlyavail
able at kosher food stores andat Rosedale Pharmacy. (See Chapters 14,
"Using Exercise to Enhance Insulin Sensitivity," and 20,"Howto Pre
vent and Correct LowBlood Sugars") Glucose tablets will not work
rapidly enough for people withsevere gastroparesis (see Chapter 22).
A liquid glucose product, Glutol, made by Paddock Laboratories,
raises blood sugars almost instantly. As with most other products
listed here, contactRosedale PharmacyifyoucannotobtainGlutol lo
cally. Dextrotabs arenot available at otherpharmacies but canbepur
chased from Rosedale or from www.rx4betterhealth.com.
MedicAlert identification bracelets. These bracelets should be
worn at all times so that if you happen to become unconscious or
confused (for example, after a motor vehicle accident or a cerebral
concussion) when you're not with a trained companion, health care
professionals willknowyou're a diabetic and takeappropriateaction.
(Although necklaces arealso available, theyareless likelythanbracelets
to benoticed in an emergency. I therefore recommend onlythe brace
lets.) Available bymail order, using forms that your physician should
be able to provide, or phone(800) 432-5378. They're also on theWeb
at www.medicalert.org. By registering with MedicAlert (the cost at
thiswritingis $45the firstyearand includes a stainless steelbraceletf)>
youcaninformthosewhotreatyouhowyouwantto betreated, which
is crucial for maintaining reasonable bloodsugars. Furthermore, you
can have your complete medical history available to EMS or emer
gencyroom personnel. This includes any directives for care you may
have in place (such as not to use a glucose intravenous drip if your
blood sugar is not too low). Phone or visit their Website for more in
*Thesize of Sweetarts varies withthepackaging, so 10mg/dlcanlikewise vary.
f MedicAlert also provides sterling silver and solid gold bracelets at additional
72 BeforeYou Start
Ketostix. These dipsticks arefor testing your urine for ketoneswhen
youarein danger of dehydration. (See Chapter 21.)
Clinistix or Diastlx. Theseare urine test strips for glucosesimilar to
Ketostix, but weusethemfor testingfood (eventhough they are mar
ketedfor testing urine). See Chapter 10, "Diet Guidelines Essential to
the Treatment ofAll Diabetics," to learn how you can use them to de
termineif packaged or restaurant foods containsugaror flour. Avail
able at most pharmacies.
Skin lubricants. In Appendix D youwill find foot careguidelines,
an importantpart of diabetic self-care. Ifyourfeet aredry, youshould
use an animal or vegetable oil lubricant. Don't use mineral oils or
petroleum-based products, asyourskin will not absorb them. Donot
use productscontaining urea or mostly glycols. Available at Rosedale
Pharmacy, drugstores, and health foodstores. Olive oil is available at
most food markets.
Bath thermometer. Many diabetics have impaired sensation in their
feet. Without knowing it, you canscald and seriously injureyour feet
if showers or baths are too hot. Don't take foot care lightly. Poor foot
care for diabetics can lead to amputation, especially if you have poor
circulation. Available at most pharmacies.
Books and publications. There are many books of food values
available that canbe helpful in tryingto figure out your meal plan (see
Chapter 11, "Creating a Customized Meal Plan"). They are optional,
but the table lists a few I think are valuable. They are all available at
most bookstores and through Internet booksellers.
Artificial sweeteners. Aswewill discussin Chapter 10,"Diet Guide
lines," the littlepackets of artificial sweetener youseeon restaurant ta
bles in the United States are predominantlyglucose, lactose, or other
sugars. Stay away frompowdered sweeteners (except stevia extract),
and always scan the lists of ingredients for any word ending in -ose.
Alsoavoid maltodextrin. In the United States, use only tablet sweeten
ers or stevialiquid or powder. You can get saccharinor Equal (aspar
tame) tabletsin anydrugstoreand in manygroceries. Stevia is sold at
health food stores. If you havea sweettooth, there is no restriction on
YourDiabetic Tool Kit 73
how many or how much of these you use. Cyclamates areavailable in
Canada and elsewhere outside the United States. These won't affect
your blood sugar.
For Insulin-Using Diabetics
Insulins. The types ofinsulins I recommend are Humalog (lispro in
sulin); Humulin Ror Novolin R(regular humaninsulin); and Lantus
(glargine insulin). For children, I recommend Levemir (detemir in
sulin), the onlylong-acting insulin currently available that canbe di
luted. These andotherinsulins arediscussed at length in Chapter 17.
You should keep at least two vials of the insulins selected byyour
doctor on hand at all times. You may need a prescription, and your
physician will select the insulin(s) appropriate for you. Athorough
discussion oftheir characteristics, use, storage, andadministration ap
pears in Chapters 16-19.
Frio. This very clever product, a wallet-style cooler activated by
immersion in water, will keep insulin cool when you are traveling
in hot climates. For periods shorter than sixweeks, insulin need not
be kept refrigerated but should not be exposed to high tempera
tures. Frio is available in the United States from Medicool, Inc.,
(800) 433-2469 or www.medicool.com/diabetes; Rosedale Pharmacy;
andwww.rx4betterhealth.com. It isavailable intheU.K. from Totally
Cool (+44-1437-741700) orwww.friouk.com.
Insulin syringes. Any 25- or 30-unit short-needle insulin syringe
with V^-unit markings should be satisfactory. They come in boxes of
100, and youshouldget at least 200 to start. Theyare available with a
prescription at most pharmacies. Their use is covered in Chapter 16,
"Insulin: The Basics of Self-Injection." Syringes with longer needles
formorerapidcorrection of elevated blood sugars shouldalso becon
sideredafter reading Chapters 16-19.
Glutose 15. You will want to showyourfriends and relatives howto
administer this glucose gel ifyouexperience confusion but not uncon
sciousness fromdangerously lowbloodsugars. Your confusion should
lift rapidly as your blood sugar increases toward thenormal range. (See
Chapter 20, "HowtoPrevent andCorrect LowBlood Sugars.")
74 Before You Start
Glucagon EmergencyKit. Ifyou don'tlive alone, it's important that
you have this for the remote possibility thatyou may become uncon
scious from dangerously lowblood sugars. You will wantto trainyour
friends, colleagues, spouse, or other family members inits use. Avail
able byprescription at most pharmacies. It's a good idea to attach a
bottle of metoclopramide syrup (below), byrubber band, to each of
your Glucagon Emergency Kits. (See Chapter 20.)
Metoclopramide syrup. Glucagon cancause nausea, anda dose of
this syrup when you regain consciousness should keep you from retch
ing.Aprescription is required.
Glutol. People with severe gastroparesis (see Chapter 22) may notbe
able to digest glucose tablets rapidly enough to bring blood sugars
back to acceptable levels ina hypoglycemic event. This glucose solu
tion is the answer. For a 150-pound diabetic, 1 teaspoon should raise
blood sugar by about 13 mg/dl (40 mg/dl per tablespoon). If your
pharmacy doesn't carry it,it's available from Rosedale Pharmacy.
How and When to Measure
Blood Sugar
The nondiabetic body is constantly measuring its levels of
blood sugar and compensating for values that are either too
high or too low.Adiabetic's body has lost much or all of this
capability. With a little help from technology, you can take over where
your body has left off and do what it once did automatically—nor
malizeyour blood sugars.
No matter how mild your diabetes may be, it is very unlikely that any
physician can tellyou howto normalizeyour blood sugarsthroughout
the day without knowing what your blood glucosevalues are around
the clock. Don't believeanyone who tellsyou otherwise. The only way
to know what your around-the-clock levels are is to monitor them
A table of blood sugar levels, with associated events (meals, exer
cise,and so on), measured at least 4 times dailyover a number of days,
is the key element in what is called a blood glucose profile. This pro
file, described in detail in the next chapter, gives you and your physi
cian or diabetes educator a glimpseof howyour medication, lifestyle,
and diet converge,and howtheyaffectyour blood sugars.Without this
information, it's impossible to come up with a treatment plan that will
normalize blood sugars. Except in emergencies, I try not to treat
someone's diabetes until I receive a blood glucoseprofilethat coversat
least one week.
Bloodglucosedata, together with information about meals, medica-
76 Before You Start
tion, exercise, and any other pertinent data that affect blood sugar, is
best recordedon the Glucograf III data sheet,illustratedon page 85.
How Frequently Are Glucose Profiles Necessary?
If your treatment includes insulininjections beforeeachmeal,your di
abetesis probablysevere enoughto render it impossible for your body
to automatically correct small deviations from a target blood glucose
range. Toachieve blood sugar normalization, it therefore maybe nec
essary for youto record bloodglucose profiles every dayfor the rest of
your life, so that you can fine-tune anyout-of-rangevalues. If you are
not treated with insulin, or if you have a very mild form of insulin-
treated diabetes, it mayonlybe necessaryto prepare blood glucosepro
files when needed for readjustment of your diet or medication.
Typically, this might be for one to two weeks prior to every routine
follow-upvisit to your physician, and for a fewweeks whileyour treat
ment plan isbeingfine-tunedfor the firsttime.Afterall,your physician
or diabetes educator cannot tell if a newregimen is workingproperly
without seeingyour bloodglucose profiles. It iswise, however, that you
also do a blood glucose profile for 1 day at least everyother week, so
you will be assuredthat things are continuing as planned.
Selecting a Blood Glucose Measuring Outfit
The measuring system usually consists of a pocket-sized electronic
meter with a liquid crystal display. The outfit will include a separate
spring-driven finger-stickingdevice and a supply of lancets. The me
ter is designed for use with disposable plastic strips, onto or into
which a drop of blood is placed. Somebrands of strips change color
when exposed to glucose, and the accompanying meter measures
color change. Most strips, however, contain electrodesthat conduct or
generatemore or lesscurrent depending upon the amount of glucose
in the blood.
About twentydifferent bloodglucose meteringoutfits are presently
being marketed in the United States. Afew of these currently have a
degree of accuracy acceptable for our purposes. Some systems rou
tinely report blood glucose values that are 40-100 percent in error.
This can be very dangerous to the user. How these have secured ap
proval from the Food and Drug Administration (FDA) is a matter of
conjecture. Usually the probleminvolves poor qualitycontrol or poor
design of the plastic strips, or inability to calibrate the meter accu
rately for different batchesof strips.
Howand When toMeasure BloodSugar 77
Although your supplier should be in a position to advise you
properly on the selection of systems for blood glucose monitor
ing, this is almost neverthe case. Even physicians and educators spe
cializing in diabetes rarely conduct the studies necessary to evaluate
these products. Reports in medical journals that purport to be evalu
ating different bloodglucose self-measurement systems are frequendy
financed by one of the manufacturers and often present grossly
deceptive conclusions. All this puts you, the consumer, in a difficult
Designs advance so rapidly that it'simpossible to predict what will
be available when you read this book. I frequently compare newme
ters for accuracy versus a major clinical lab. I also checkthe repro
ducibility of results. You can call our Diabetes Center at (914)
698-7525 Monday through Thursdaybetween9:30 a.m. and 3:00 p.m.
Eastern (U.S.) time to findout what systemwe currently recommend
for our patients.
There are two things to look for in a meter: accuracy and repro
ducibility (precision). Other"features" are nothingmorethanmarket
ing gimmicks. You want a meter that if you were to take several
readings, oneimmediately after theother, wouldgive youthe same re
sults. Themeters that I recommend are selected with precision andac
curacy in mind. Buy from adealer whowill refundyour money if you
find the systemto be inaccurate. You can test the meter right in the
store by taking four readings in succession. They should be within 5
percent of one another when blood sugars are within the 70-120
mg/dl range. I have not found anymeters that are suitably precise or
accurate above 200 mg/dl. Ask your physician about the systems he
has evaluated. He cansecure virtually anysystem from its manufac
turer for study at no cost.
A number of my patients have been tempted by advertising for
bloodsugar meters that contain abuilt-indevice to puncture the skin
at sitesother than the fingertips (arms, buttocks, abdomen), wherethe
puncturecauses absolutely no pain. I have testedseveral oftheseprod
uctsand found theirbloodsugar readings to be inaccurate. Far supe
rior to these is the Bayer Vaculance for painless punctures of these
alternate sites. I use it myself and a separate, very accurate meter. If
bloodsugar ischanging rapidly, these alternate testsite results maylag
behind fingertip tests by as much as 20 minutes. In my experience,
nearly all of the finger sticks thatI perform onanyone are painless be
cause I use the technique described in the following section.
78 BeforeYou Start
Manyinstruction bookletsgive inadequate or erroneous instructions
for preparing and pricking the finger or putting the drop of blood
onto or into the test strip or electrode. If the instructions that follow
conflict with what you've been told, believe mine. My techniques
aren't based on something I read in medical school or in a medical
journal. They're the ones I use on myselfeveryday. I've been measur
ing my own blood sugars for more than thirty-five years and have
performed hundreds of thousands of finger sticks on myself and
thousands of my patients.
1. If you've handled glucose tablets, skin lotion, or any food since
last washingyour hands, washthem again. Invisible material on
your fingers can cause erroneously highreadings. Certainlywash
your hands if they are soiled. If you are sitting in a car or some
other placewhereyou cannot washyour hands, lickthe appro
priate finger enthusiastically and dry it on a handkerchief or
clothing. Don't wipe your fingers with alcohol; this will dry out
the skin and can eventuallyfoster the formation of calluses. Nei
ther I nor any of my patients have developed finger infections by
not using alcohol.
2. Unless your fingers arealreadywarm, it maybe necessaryto rinse
them under warm water. Blood will flow much more readily
from a warm finger. If outdoors in cold weather, store your me
ter in a pocket next to your body and put your finger under your
tongue to warm it.
3. Layout all the suppliesyou will need at your work area. These
usually include a finger-stick device loaded with a lancet, your
blood glucose meter, a blood glucosetest strip, and a tissue for
blotting your finger after the test. If youhaveno tissue,just suck
off the blood (unless your religion forbids consuming human
blood). Insert a disposabletest strip into your blood sugar meter.
(Some test strips are supplied in individual foil packets. The ac
companying instruction booklet may tell you not to handle the
strips directly but to hold them in the foil. This assumes that
your hands are always dirty, which is absurd, since they must be
clean for an accurate result.)
Howand When toMeasure BloodSugar 79
4. Many spring-activated finger-stick devices come with two rigid
plastic covers for the end that touches your finger. Usually one
cover is for thin or soft skin (as in small children), while the other
is forthick or callused skin. To get a shallower puncture, use the
thicker-tipped cover; to get a deeper puncture, use the thinner-
tipped cover. Most finger-stick devices havearotarycontrol that
can be dialed to the depth of the puncture that you prefer. An
evendeeper puncturemaybe obtained by strongly pressing your
ringer against the lancet cover. A very shallow puncturemay be
obtained by barely touching the fingertip to the cover. The pres
sureof the finger on the cover determines howdeepthe puncture
will go. It shouldbe deepenoughto provide anadequate dropof
blood, but not be sodeep as to cause bruising or pain. Contrary
to common teaching, the best sites for prickingfingers are actu
allyon the back of the hand. Prick your finger betweenthe first
joint and the nail, or between the first and second joints (not
overthe knuckles), as shownby the shaded areas in Figure 4-1.
Pricking these sites should be less likelyto cause pain and more
likelyto produce adropof bloodthanwill pricking your fingers
on the palmar sideof the hand. You will alsobe free from the cal
luses that occur after repeated punctures on the palmarsurface
Fig.4-1. Sites toprick on thedorsum
Fig.4-2. Sites toprick onthepalmar
surface ofyourfingers.
80 BeforeYou Start
of the fingers.* When using this technique, I press the tip of the
lancingdevice verygently against the finger, asthe skin is thinner
there than on the palmar surface. If you find it repugnant to
prick the dorsum (knuckle side) of your fingers, use the sites on
the palmarsurface illustrated in Figure 4-2. Personally, I actually
use all of the sites shown in both diagrams. As you will not be
sharing your finger-stick device, you need not discard the dis
posable plastic lancets with the metal point after every finger
stick. It is a good ideato discard them once a month, as they do
eventuallybecome dull.
5. Over a period of time, you should use all the fingers of both
hands. There is no reason to prefer one finger over the others.
Onceyou havepricked your finger, squeeze it (usearhythmic ac
tion rather than steady pressure) with the opposite hand until
the drop of blood is about Vio inch (2 mm) in diameter. As you
squeeze, the index finger on your squeezinghand should be be
hind the distal (outermost) joint of the finger you aresqueezing.
If flow is inadequate (see items7 and8), performadeeper finger
6. Touch the dropof bloodto the proper point on the test strips
7. Most meters will start an automatic countdown as soon as the
strip has absorbed enough blood. The countdown, in seconds,
usuallyappears on the displayscreenand concludeswith the ap
pearance of your blood glucose value. If your meter has a timer
button, pressit immediately after applying blood, without delay;
do not stop to examine the strip in order to determine whether
or not you haveappliedenough blood. (This comes later.) Since
the accuracy of the test usually depends upon the timing, the de-
* My patients and I are much indebted to Mr. Ron Raab, president of Insulin for
Life, Inc., of Caulfield, Australia, for this not-so-obvious technique. Mr. Raab's
attempts to publish this important findingwererepeatedlyscorned by medical
journals and finallycameto myattention via personal correspondence. I use Mr.
Raab'stechnique myselfand find it far superior to the palmar technique, which I
used for years. Bythe way, this techniquewasin common useby physicians sev
enty yearsago.Like so manythings in medicine, it had to be rediscovered.
+Most manufacturers providestripsthat havean invisible holeat their tip. Blood
is sucked into the strip by capillaryaction. The puncture site should point up
ward and the tip of the strip must be inserted into the drop of blood. With such
strips, the blood should not be put on top of the strip as erroneouslypracticed by
many users.
Howand Whento Measure BloodSugar 81
laybetween applying the blood and pressing the button should
be no greater than 1second. This doesn't sound like much time,
but you'll be an old pro in short order.
8. Most meters have an automatic timer that begins a countdown
preceded by an audible beep when the strip has been filled.
Sometimesthe beep occursbeforethe strip is full, so it is wiseto
leave the tip of the strip in thebloodfor 2seconds after the beep.
After the 2 seconds have elapsed, youmay examine the strip to
makesure it is adequatelycovered or filled with blood. If it is not,
discardthe strip and start again.
9. If youget a little blood onyourclothing, rubon some hydrogen
peroxide witha handkerchief. Wait forthe foaming to stop.Then
blot and repeat the process. Continue until blood has disap
peared. This works best while the blood is still wet.
10. When your meter finishes its countdown (or countup, depend
ingon the model), yourbloodsugar will beshownon the display
screen. Write it down on your Glucograp III data sheet as in
structed in Chapter 5.
11. If you are measuring someone else's blood sugars using your
personal equipment (not a wise practice), install a fresh lancet
each time, and wipe off the end cap of the finger-stick device
with fresh bleach after each use. It is possible to transmit seri
ous infectious diseases from one person to another via finger
Theentire process, from pricking thefinger toa final reading, takes
as little as 5 seconds, and rarelymore than 30seconds.
Note: Donot expect accurate blood sugars (orHgbAlc) ifyou have
been takingmore than 250 mg per dayof a vitamin C supplement.
Readings will be lower than the true values.
Make sure you have all the suppUes you and your physician have
checked off in Chapter 3. Put a stringon your finger to remind youto
asksomeone at the doctor's office to watch you measure your blood
sugar and to correct any errors you may make. (About 80 percent of
mynewpatients arenot measuringtheir bloodsugarsaccuratelywhen
82 BeforeYou Start
I first seethem.) Bringalongat least two weeks' worth of your blood
glucose profiles. Ideally, these should be written on Glucograf III
datasheets (seenext chapter),which havebeen designed for quick re
view by the physician or other health care professional. To compile
your profiles, blood sugars shouldbe measured:
• Upon rising in the morning
• Immediatelybeforebreakfast
• Five hours after every injection of rapid-acting insulin (if you
use one of thesebeforemealsor to coverelevated blood sugars);
otherwise, before each meal
• Two hours after meals and snacks
• At bedtime
• Beforeand after exercising, shopping, or running errands
• Whenever you are hungry or suspect that your blood glucose
may be higheror lowerthan usual
• Before driving a car or operating heavy machineryand hourly
while engagingin these activities
Once your blood sugars have been fine-tuned, it may not be neces
saryfor you to checkthem2hoursafter meals andimmediatelybefore
breakfast. Appropriatetimes fortestingwill be specified in subsequent
Recording Blood Sugar Data
Your blood sugar levels are affected by avarietyof things: what
medications you are taking (such as insulin or oral hypo
glycemic or insulin-sensitizing agents), what exercise you may
haveperformed,whether you'vegot aninfectionor cold,what you ate,
when you ate itl and others you will discover as you track your blood
sugars. Many people find, for example, that such things as public
speaking or arguments will raise bloodsugars. All of thesebits of in
formation — not just your blood sugar levels — need to be recorded
and taken into account. Without this detailed information— your
own personal Mood sugar profile — your physician or diabetes edu
cator cannot assist you in developing an ongoing program for blood
sugar normalization. To my knowledge, none of the many forms or
computer programs currently available for this purpose show ade
quateinformation in areadily usable format. The Glucograf III data
sheet,* likeour program, was designed byadiabetic engineer (me) for
Glucograf III data sheets are printed identicallyon both sides so
that each page provides space for two weeks' worth of data. If your
*Glucograf is a registeredtrademark ownedbyRichardK. Bernstein,MD. The
data sheet form is protected by U.S. copyright, and may not be reproduced for
sale without permission of the author. Readers of this book who wish to have
some practice copies for immediate use may make photocopies of the form,
whichis reproduced at a reducedsizeon page85in order to make it fit into this
book Any photocopying shop can enlarge the image in order to provide you
with standard 8V4 x 11sheets. Padscontainingenough pages to cover one year
canbe ordered byphone from Rosedale Pharmacy, (888) 796-3348.
84 BeforeYou Start
physician wants detailed information about the content of each of
your meals, use one side to list meal content and the reverse to list
medication,blood sugars, exercise, the timesof your meals, and so on.
The data sheet is designed so that youcan foldit up and carry it with
you. I recommend carrying a fine-point pen (0.1 mm) with you as
well. It will helpwhenspace is tight —whichis likely, particularlyin
the medication, exercise, food, etc. column, where much informa
tion must be written in a small space. If youwill be faxing your sheets
to your physician, do not use pencil, as it doesn't always transmit
The rest of this chapter is divided into sections corresponding to
column and field headings on the Glucograf III form, and explains
the sortsof thingsyouought to berecording and the most informative
ways for doing so.
Across the top of the datasheet, thereareseveral fields with space for
entering important information.
name. Entering your name will ensure that the form will end up in
your chart at your doctor's office and not in someone else's.
doctor's phone. This fieldshould contain the telephone number at
which you can reach your physician when you are asked to discuss
your blood sugar and other data.
doctor's fax. If youwill befaxing your datasheets, enter your physi
cian's faxnumber as well. Alternatively, if you would usuallyscan and
e-mail your form, you might enter his e-mail address here — or just
put it in your "address book."
target bg. This is the blood sugar goal that your physicianwill assign
and that you will try to maintain. Althoughnormal is approximately
80-90 mg/dl, in certaininstances your physician mayopt for a higher
value for a brief period. If you'veendured veryhigh blood sugar levels
for an extendedperiod of time, your physician willnot instantlytry to
normalize your blood sugars, as you may at first feel uncomfortable
(hypoglycemic) at a normal value. If you take insulin, he'll assign a

86 Before You Start
series of intermediate target values, together with instructions for
correctingbloodsugars to reach theselevels asyouwork towardblood
sugar normalization. If your initial blood sugars showthat you are in
the 300-400 mg/dl range, he might set atarget of, say, 175 mg/dl fora
brief time. If you have gastroparesis (delayed stomach-emptying; see
Chapter 22) and use insulin, you are at real risk for severe hypo
glycemia. Your physician may therefore recommend a target well
abovenormal for an indefiniteperiodof time, to provideasafety fac
tor that reducesthe likelihood of very lowblood sugars.
usual doses of insulin or oral agent. If you require insulin or an
insulin-sensitizing or insulin-mimetic agent* to maintain your target
range, you will have to follow a precise regimen. It will therefore be
important to have your blood sugar medications spelled out so that
evenif you forget, you can always refer to the doses andtimes in this
field. When you are to take a medication before a meal, cross out
"post" (as shown below); whenyouare to takeone after ameal, cross
out "pre."
If yourphysician asks youto change the dose of oneof yourblood
sugar-lowering medications, put aline through the prior dosage and
enter the new dose to the right of the old one, as in the following ex
Upon Arising £4=AH l^UAN
Min. pre/post bkfst.
Min. pre/post lunch.
10 Min. pre /^dinner %X CflOHeJ- 2 X ZOO Kef
Min. pre/post snacks
At Bedtime
In this fictitious example, the patient hadbeeninjecting 2 units of
Lantusinsulinon arising when this data sheetwasstarted. In addition
to insulin, he had been takingthree 500mg tablets of metformin (an
insulin-sensitizing agent) 90minutes before dinner. During the week
the sheet covers, his doseof insulin on arisingwas reducedto \Vi units
and his dose of metformin before dinner was reduced to two 500 mg
tablets. Retaining the olddoses in this field cangive your physician an
importantat-a-glance history of the changes that were made.
* Insulin-sensitizing and insulin-mimetic agents arebloodsugar-lowering pills
that youmaybe using. Theyarediscussed indetail in Chapter15.
RecordingBloodSugar Data 87
l unit will lower blood sugar [insert the abbreviation of
your most-rapid-acting insulin from upper right-hand corner of the
Glucograf form] .This field is for useonlyby people who takeinsulin
anduse rapid-acting insulin to bring down elevated blood sugars. In
Chapter 19, we'll discuss guidelines for calibrating the effect that 1
unit of rapid-acting insulin will have upon yourblood sugar. Mean
while, enter on the form the amount of blood sugar reduction that
your physician suggests willbe achieved by injecting 1unit.
miscellaneous. This field (box) is for anyother pertinent guidelines
orinstructions that youmayhave difficulty recalling. Some people en
ter the times they should check their blood sugars. Thus, depending
on your regimen, you might write:
/ BG -on arising -bedtime
-before meals -whenhungry
-2 hr post meals
If this field istoo small for all youwishto enter, usethe top margin
of the form.
bg effects of sweets. If you use insulin or oral agents, you will be
taughthowto use glucose tablets to raise yourblood sugar rapidly. In
Chapter 20, we'll discuss howyou will calibrate the effect that 1tablet
has on yourblood sugar. Thus, if 1Dextrotab raises your blood sugar
8 mg/dl, you would write:
Alternately, if your brand of glucose tablet isWackyWafers (which
might raise your blood sugar 10mg/dl), you could write:
Youwill also learnto calibrate the effect 1gramof carbohydrate has
onyourbloodsugar. If 1gramwillraise yourbloodsugar 5mg/dl, you
would write:
1gmCHO -> T5
exercise adjustments. This fieldis alsoused only if you use insulin
or oral agents. It reminds you what to eat forvarious forms of exercise
to prevent your blood sugar fromdroppingtoo low.Thus, if you were
88 Before You Start
planning to spend the afternoon at a shopping mall (which can be
treacherous, because this often requires considerably more walking
than we realize), you may be advised to eat half a slice of breadat the
start of every hour to keep your blood sugar from falling too low.
Thus, you might write:
Mall — y2brd/hr
abbreviations. Space constraints make it necessary to use abbrevia
tions. You may be tempted to use your own, but in order to avoid con
fusion, I recommend usingthe short list of standardizedabbreviations
provided in the top right-hand cornerof the data sheet. Using these
abbreviations will helpboth you andyour physician to know immedi
atelythe details of"events" that affect your blood sugar.
As you can see, each day is broken up horizontally into three
columns — time; blood sugar; and medication, exercise, food,
etc. Vertically, each column is broken up into 3-hour blocks, ex
cept the 9pmthru 1amand l amthru 6amblocks, which are4-hour
and 5-hour blocks, respectively. During each day, you will experi
ence various "events" involving your blood sugar. An event may be
a meal, a dose of medication, exercise, or even a blood sugar mea
surement itself. These should be recordedin the corresponding col
umn and time block. You should not record a dose of medication that
does not affect your blood sugarlevels, such as blood pressuremed
time. In this column, write the exact time of the event. If you mea
sured your blood sugar at 1:30 p.m. on Tuesday, write 1:30 in the 12
noon thru 3 pmblock of the time column forTuesday.
blood sugar. In this column, write all blood sugar readings. If for
some reasonyou do not haveyour blood sugarmeter with you (a mi
nor crime) and you experience symptoms suggestive of low blood
sugar, write "low" in this column in the appropriate time block
and proceed with the instructions for correcting low blood sugar in
Chapter 20.
RecordingBloodSugar Data
medication, exercise, food, etc. This column is a catchall where
youshould record allevents otherthanblood sugar readings. Follow
ing are a few examples of events and howyou would recordthem in
abbreviated form in the proper block:
Injected 5 units of Lantus insulin
Ate breakfast
Consumed more food at dinner than prescribed Tdin
Took 3 Dextrotabs (glucose tablets) 3DT
Tooktwo 500mg Glucophage (metformin) pills
2x500 MET
Walked 2 miles
Went shopping for 3 hours
Shop 3 hr
Injected\Vi units Humaloginsulin,
intramuscularly (into a muscle)
Sorethroat all day
Sore throat [enter
at the top of the
day's column]
Went to dentist
Onceyour bloodsugars have beenfine-tuned on one of the regimens
described in this book,weexpect that theywill remainwithinnarrow
limits of your target value most of the time. There will, in all likeli
hood,beinstances when yourblood sugars will deviate from yourtar
get range.
ShowWhat Caused Blood Sugars to Deviate
Sometimes you maystickprecisely to your diet and medicationplan
but then find yourself in a restaurant andsimply incapable of letting
the dessert cart go by without partaking of its wonders. Your blood
sugars will naturallyshowa precipitous rise. Or youmayget someex
ercise that makes bloodsugargotoo low. To makeit easyfor both you
and your physician to understand and evaluate such connections, cir
clethe cause, then circle the resulting bloodsugarvalue, and connect
the two circles with a line. For example, a high morning blood sugar
90 BeforeYou Start
might be circled and connected to "snack" at bedtime the previous
Circle Puzzling Blood Sugar Values
Even though you stickto your regimen with an iron will, your data
will sometimes show an unexpectedlyhigh or lowblood sugar value.
Repeat the measurement after washing your hands (to remove any
traces of food or glucose) and ensure that you haven't inadvertently
slipped on your measurement technique. If the unexpected reading
persists, circle thisvalue, asit mayrequire further investigation. There
are several strange biologic phenomena that can afreet your blood
sugar, and these aredetailed in the next chapter. Your physician or di
abetes educator should help you figure out the cause of unexpected
blood sugar readings sothat youcanpreventor anticipatethem in the
Nowthat you have beenexposed to bloodsugarself-monitoring and
the recording of data,youcanbegin using this knowledge to normal
izeyour blood sugars.
Strange Biology
Sometimes, even whenyouthinkyou're doingeverything right,
your blood sugars maynot respond as you expect. Often this
will be due to one or moreof thebiologic curiosities that affect
diabetics. The purpose of this chapter is to acquaint you with some
realphenomenathat canconfoundyour plans, but whichyoucanfre-
quendy circumvent if you are aware of them.
Figure 1-2, page 45, illustrates the normal, nondiabetic blood insulin
response to a meal containing carbohydrate and protein. When glu
cose from dietary carbohydrate enters the bloodstream, beta cells of
the pancreasrespond — or should respond — immediatelyby releas
ing stored insulin granules. These granules may have been stored for
many hours in anticipation of what is known as a glucose challenge.
This rapid releaseis calledphase I insulin response.
The nondiabetic body will utilize this immediate releaseof insulin
to prevent blood sugar from increasingsignificandy. As we discussed
in Chapter 1, one of the hallmarks of type 2 diabetes is the diminished
ability to do this. Therefore, blood sugars will shoot up after eating
(carbohydrates in particular) and will be brought back into line only
slowlybyphase II insulin response(the release of newlymanufactured
insulin). This blood sugar rise can be minimized,primarily by dietary
manipulation, but for some diabeticsby diet and/or oral agents or in
jected insulin.
A possible but unproven explanation for diminished or absent
92 Before You Start
phaseI insulinresponse in diabetics isthat the beta cells are still capa
ble of making insulin but not capable of storing it. In this model, in
sulin would be releasedalmost as soon as it is made. This inability to
store insulin could also explain the inappropriate release of insulin
that often occurs when blood sugar is alreadylowin very early type 2
diabetes. Such individuals may experienceblood sugars that are both
too high and too lowin the sameday—even without medication.An
alternate explanationis that the sensitivity of the beta cells to changes
in the blood sugar diminishes, so that they respond inadequately to
such changes.
Youmay begin to notice as you regularlymonitor your blood sugars
that your fasting blood glucoseon waking in the morning is consider
ablyhigher than it waswhen you went to bed, eventhough you didn't
get up for a midnight snack.There are three common causesfor this:
gluconeogenesis, the dawn phenomenon, and gastroparesis (delayed
Gluconeogenesis, which we discussed briefly in Chapter 1, is the
mechanism by which the Uver (and, to a lesser degree, the kidneys and
intestines) converts amino acids into glucose. Dietary protein is not
the only source of amino acids. The proteins of your muscles and
other tissuescontinuallyreceive amino acids fromand return them to
the bloodstream. This constant flux ensures that amino acids are al
ways available in the blood for conversion to glucose (gluconeogene
sis) by the liver or to protein by the muscles and vital organs. Some
diabetics still make adequate insulin to prevent gluconeogenesis.
However, once your insulin production drops below a certain level,
your liver (and your kidneys and intestines) will inappropriately pro
duceglucose and thus raise your bloodsugarevenwhileyou'refasting.
In all likelihood, you won't be ableto control this phenomenon by
diet alone, particularlyifyou'reatype 1diabeticor a type2makingfar
too little insulin to offsetyour insulin resistance. For type 2s, appro
priate weight loss and vigorous exercise may be most helpful in im-
StrangeBiology: PhenomenaPeculiartoDiabetes 93
provingthe sensitivityof the liverand muscles to whateverinsulin re
mains. The most reliable treatments will involve medication, either
certain oral agents or insulin. If you're obese, however, large doses of
insulin can make you more obese and more resistant to insulin. So a
majorgoal shouldbe to bringyourweight intoline.
The Dawn Phenomenon
As youknow, I'ma type 1diabetic. I may nolonger make anyinsulin
at all. If I decide tofast for24hours —eatabsolutely nothing —I will
need to inject 2% units of long-acting insulin in the morning to pre
vent gluconeogenesis for 18 hours. IfI check myblood sugar everyfew
hours, it will remain constant, confirming thattheinsulin issuppress
ing gluconeogenesis.
If, 18 hoursafter myfirst injection —andwhile still fasting —I in
ject another 2% units of insulin, common sense would maintain that
thissecond dose should suppress gluconeogenesis overnight.
SoI gotosleep andawaken 9-10hours later. Onarising, I check my
blood sugar. Instead of being constant, as it was during mywaking
hours, it's now20-100mg/dl higher than it was at bedtime.
If I were to try the same experiment a week later, I'd experience
about the sameovernight risein bloodsugar. Why?
Although the mechanics of the dawn phenomenon aren't yet en
tirelyclear, research suggests that thefiver deactivates morecirculating
insulinduring the early morninghoursthan at other timesof the day.
It doesn't matterwhether youmade theinsulin yourself or injected it;
theliver hasno preference. Withinadequate circulating insulinto pre
vent gluconeogenesis, your bloodsugars maybe higher in the morn
ingthan theywereat bedtime.* Thisisn't a problemfor a nondiabetic,
because a body with fully functional pancreatic beta cells will just
make more insulin.
Investigators have actually measured blood sugar every hour
throughout the night under similar circumstances. Theyfindthat the
entire blood sugar increaseoccursabout 8-10 hours after bedtime for
mostpeople whoaresoaffected. Thatdoesn't mean, however, that you
should sleep only 7 hours a night to try to avoid it. Both the time it
takes for bloodsugar to increase and the amount of the increase vary
*Consuming alcohol at bedtimecaninhibit gluconeogenesis overnight, but not
in a predictable fashion.
94 BeforeYou Start
from one person to another. An increase may be negligible in some
and profound in others. This is one of many reasons why any truly
workableprogramfor blood sugar normalizationmust be tailored to
the individual.
Though it is more apparent in type 1diabetics, many type 2 diabet
ics also show signs of the dawn phenomenon. As you will see, the
treatments described in this book enable us to circumvent this blood
sugar rise.
This condition has a chapter all its own (Chapter 22), and we will dis
cuss it there in detail. However, it's important to mention it in anylist
of factors that can lead to puzzlingblood sugar readings.
Most people who've had long-standing diabetes develop some de
gree of damageto the nerves that govern the muscles of the stomach
and intestines. Gastroparesis diabeticorum (the weak or paralyzed
stomachof diabetics) is caused by manyyears of elevated blood sug
ars. If you're a type 1,or a type 2whoisn't makingsignificantamounts
of insulin, it can haveunpredictableeffects on blood sugar.
Like diabetes itself, gastroparesis can be mild to severe. In extreme
cases, people maywalkaround for dayswith constipation, belchingor
vomiting, midchest burning, and bulgingstomachs.Much more com
mon, however, is mild gastroparesis in which physical symptoms are
not apparent but blood sugarsare erratic.
The big problemswith gastroparesis ariseif you'retakinginsulin. If
you take your insulin before a meal to prevent a subsequent rise in
blood sugar but the meal remains in your stomach and glucose doesn't
enter the bloodstream as predicted, the insulin can take your blood
sugardangerously low. I knowthreeindividuals whoexperienced daily
episodes of unconsciousness and seizures fromtime to time aftermeals
for several years before I met them and diagnosed this condition.
Thereare, however, ways of greatiy improving bloodsugarsin spite
of the unpredictabilityof this condition, and these are discussed in
Chapter 22,"Delayed Stomach-Emptying."
Sustained Emotional Stress
For years,manyphysicians havebeenblamingemotional stressfor the
frequent unexplained blood sugar variations that many patients expe-
Strange Biology. Phenomena Peculiar to Diabetes 95
rience. This is an evasive and possibly self-serving diagnosis. It puts
theresponsibility for unexplained variations inblood sugar onthepa
tient's shoulders andleaves thephysician with noobligation toexam
ine the treatment regimen. Certainly there is no question that stress
canhave adverse effects upon your health. I have reviewed more than
a million blood sugar entries from many patients, including myself.
Onecommon feature ofallthis dataisthat most prolonged emotional
stress rarely has a direct effect uponblood sugar. This kind of stress
can, however, have a secondary effect by precipitating overeating,
binge eating, or indulgence in kinds of eating that will increase blood
I knowmany diabetics who've been involved instressful marriages,
divorces, lossof a business,slowdeath of a closerelative, and the count
less other sustained stresses of life we all must endure. These stresses
have one thing in common: they aren't sudden but usually last days,
or even years. I haveyet to see such a situation directly causeblood
sugar to increase — or, for that matter, decrease. An important thing
to remember duringsustained periods of life when everything seems
out of control is that at least you can control one thing: your blood
Adrenaline Surges
Many patients have reported sudden blood sugar spurts after brief
episodes of severestress. Exampleshave included an automobile acci
dent without physical injury; speaking in front of a large audience;
taking veryimportant exams in school; and having arguments that
nearly become violent. I am occasionally interviewed on television,
and I always check —and, if necessary, adjust —mybloodsugarim
mediately before and after such appearances. Until I eventually be
came accustomed to such appearances, my blood sugar would
inevitably increase 75-100 mg/dl, even though on the surface I might
haveappeared relaxed. As a rule of thumb, from personal experience
and from observingmy patients, I wouldsaythat if an acute event is
stressful enoughto start your epinephrine (adrenaline) flowing, as in
dicated byrapid heart rateand tremors, it is likely to raiseyour blood
sugar. Epinephrine is one of the counterregulatory hormones that
cause the liver to convert stored glycogen to glucose. This is part of
what is often calledthe "fight or flight" response, your body's attempt
to provideyouwithenoughextraenergyeitherto overcome an enemy
or run likeheckto get away. Type2 diabetics who makea lot of insulin
96 BeforeYou Start
are less likely to have their blood sugar reflect acute stress than are
those who make little or none.
An occasional bloodsugar increase after a verystressful event may
well have been brought on by the event. On the other hand, unex
plained blood sugar increases extending for days or weeks can rarely
be properly attributed to stress. I know of no instances where pro
longed emotional stress caused abnormal blood sugars in diabetic or
nondiabetic individuals.Therefore,if you experiencea prolonged un
explained change in your bloodsugarlevels after extended periodsof
normal blood sugars, it is wise to seek out a cause other than emo
tional stress.
General Anesthesia
If not treated with specialdosingof insulin, type 1and most type 2 di
abetics with previouslylevel, normal blood sugars may experience a
blood sugar increase during surgery that is accompanied by general
Insulin Resistance Caused by Elevated Blood Sugars
Thereare at least five causes of insulin resistance —inheritance, dehy
dration, infection, obesity, and high blood sugars. Insulin's ability to fa
cilitatethe transport of glucose fromthe bloodinto Uver, muscle, fat,and
other cells is impairedasblood sugarrises. This reducedeffectiveness of
insulin, known as insulinresistance, has been attributed to a phenome
non called postreceptor defects in glucose utilization. If, for example, 1
unit of injectedor self-made insulin will lower your blood sugar from
130 to 90 mg/dl, someone with insulin resistance caused by elevated
bloodsugars mayrequire3 unitsto lower it from430to 390mg/dl.
Considerwhat might happenif I, a type 1diabetic, am fasting and
inject just enough long-acting insulin to keep my blood sugar at 90
mg/dl for 18 hours. If I eat 8 grams of glucose — enough to raise my
blood sugar to 130mg/dl — the chances are that, because of the ele
vatedblood sugar, mybloodsugarwon't just riseto 130mg/dl and re
main there. It will continue to riseslowly throughout the day, so that
12hours after I consumedthe glucose, myblood sugar might actually
be 165mg/dl. Insulin resistance, at least for type 1diabetics, occurs as
blood sugar increases, and soelevatedblood sugar should be corrected
as soon as it's feasible. Delay willonlypermit it to risehigher. Because
type 2s still produce some insulin, their bodies are more likelyto cor
rect the blood sugar rise automatically.
Strange Biology: Phenomena Peculiar toDiabetes 97
We will discuss dehydration as a cause of insulin resistance in
Chapter 21. Infections are discussed attheendof thischapter andalso
in Chapter 21.
Manyyears ago apatient asked me whyherbloodsugar went from90
mg/dl up to 300 mg/dl every afternoon after she went swimming. I
askedwhat she ate beforetheswim. "Nothing, just afreebie," she replied.
As it turned out, the "freebie" was lettuce. When I asked her just how
much lettuce shewas eating before her swims, shereplied, "Ahead."
A head of lettuce contains about 10 grams of carbohydrate, which
canraise atype 1adult's bloodsugar about 50mg/dl at most. Sowhat
accounts for the other 160 mg/dl rise in herbloodsugar?
The explanation lies in what I call the Chinese restauranteffect. Of
ten Chinese restaurant meals contain large amounts of protein or
slow-acting, low-carbohydrate foods, such as bean sprouts, bok choy,
mushrooms, bambooshoots, andwater chestnuts, that canmakeyou
feel full.
How canthese low-carbohydrate foods affect blood sugar so dra
The upper part of the smallintestinecontains cells that release hor
mones into the bloodstream when they arestretched, as after a meal.
These hormones signal the pancreas to produce someinsulin to pre
vent the blood sugar rise that might otherwise follow the digestion of
ameal. Large mealswill cause greater stretching of the intestinalcells,
which in turn will secrete proportionately larger amounts of these
hormones. Since averysmall amount of insulin released by the pan
creas cancause a large drop in blood sugar, the pancreas simultane
ously produces the less potent hormone glucagon to offset the
potential excess effect of the insulin. If you're diabetic and deficient
in producing insulin, you might not be able to release insulin, but
you will still release glucagon, whichwill cause gluconeogenesis and
glycogenosis and thereby raise your blood sugar. Thus, if you eat
enoughto feel stuffed, yourbloodsugar can goup by alarge amount,
even if you eat something undigestible, such as sawdust. Even a small
amount of anindigestible substance will cause ablood sugar increase
in type 1diabetics if not covered by an insulininjection.
Complicatingmatters further, pancreatic betacells alsomake ahor-
98 Before You Start
mone called amylin. Amylin inhibitsthe effectiveness of glucagonand
works on the brain to causesatiety. It alsoslows stomach-emptying to
discourage overeating. Withfew or no betacells, diabetics don't make
enough amylin, and consequentiy theytend to remainhungry after
eating and show an exaggerated Chinese restaurant effect. Since the
last edition of this book, amylinsubstituteshavebecomeavailable and
have found an important use in the prevention of overeating (see
The first lesson here is: Don't stuffyourself. The second lesson is:
There's nosuch thingasafreebie.* Anysolidfood that you eat can raise
your bloodsugar.* If youcan't controlyour overeating, seepage 247.
Exercise can havevaryingeffects upon blood sugar, depending upon a
number of variables, including the type of exercise, howvigorously it's
performed, whenit isperformed, andwhattypeof medication youare
using, ifany. These effects aretoovaried andnumerous todiscuss inthis
briefspace. Please see Chapter 14, "Using Exercise to Enhance Insulin
Sensitivity," if you are embarking on an exercise programor findyour
bloodsugars unpredictably affected byyour existing exercise program.
At the time they are diagnosed, type 1 diabetics usually have experi
encedveryhigh blood sugars that cause a host of unpleasant symp
toms, such as weight loss, frequent urination, and severe thirst. These
symptoms subsidesoon after treatment with injectedinsulin begins.
After a fewweeks of insulin therapy, many patients experience a dra
matic reduction of insulinrequirements,almost asif the diabeteswere
reversing. Blood sugars maybecome nearly normal,even withdiscon
tinuation of insulin injections.This benign"honeymoon period" may
* Except for noncaloricfluids that flow through the intestines without causing
t Several readers from China have e-mailed me that their restaurants don't use
sweet saucesso this effectshouldn't applyto them. This is a misunderstanding of
this effect— the Chineserestaurant effectis causedby any solid foods.
StrangeBiology: PhenomenaPeculiar to Diabetes 99
last weeks, months, or even as long as ayear. If themedical treatment
isconventional, thehoneymoon period eventually terminates andthe
well-known roller coaster of blood sugar swings ensues.
Why doesn't the honeymoon period last forever? My experience
with patients indicates thatit can, with proper treatment. But there are
several likelyreasons why it doesnot with conventional treatment. At
thiswriting, however, they still remain speculative.
• The normal human pancreas contains many more insulin-
producing beta cells than are necessary for maintaining normal
blood sugars. For blood sugar toincrease abnormally, atleast 80
percent of thebeta cells musthave been destroyed. Inearly type
1 diabetes, many of the remaining 20 percent have been weak
ened byglucose toxicity from constant high blood sugars and by
overwork. These betacells can recover if theyare given arest with
thehelp of injected insulin. Even if they recover, however, they
still must workat least five times as hard to match the jobof a
normal pancreas working at 100 percent capacity. Eventually,
withconventional treatment, this overwork helps cause themto
break down.
• It is nowbelieved that highbloodglucose levels are toxicto beta
cells. Even abriefbloodsugar increase after ahigh-carbohydrate
meal maytake asmall toll. Over time, thecumulative effect may
wipe them out completely.
• The autoimmuneattack upon beta cells, the presumed cause of
type 1diabetes, is focused upon several proteins. One is insulin,
and another is present on the special vesicles — or bubbles —
that are formed at the outer membrane of the beta cell. These
vesicles containinsulin. Normally, theyburst at the surface ofthe
cell, releasing insulin granules into the bloodstream. The more
vesicles created when more insulin is manufactured, the greater
the autoimmune attack upon the beta cell. If less insulin is re
leased, lessof this proteinis exposedto attack.
Based uponmy experience withthe fair numberof type 1diabetics
I've treated from thetimeof diagnosis, I'mconvinced thatthe honey
moon period can be prolonged indefinitely. The trick is to assist the
pancreas andkeep it asquiescentaspossible. With the meticulous use
of small doses of injected insulin and withthe essential use of avery
lowcarbohydrate diet, the remaining capacity of the pancreas, I be
lieve, canbe preserved.
100 Before You Start
Another kind of stress to which your body can become subject —
and which can muddy and in some instances wreak havoc on your
best efforts to control blood sugars — is infection. I have saved this
category of stress for last not because it isthe leastimportant, but be
cause,when present, it can be the most important.
A kidney infection, for example, can triple insulin requirements
overnight. When blood sugar rises unexpectedlyafter weeks of normal
values, it is wiseto suspectinfection. I havenoted that my own blood
sugars rise24hours before the onsetof a sorethroat or cold.Everyone
in myfamily takes Sambucol (an extract of the black elderberry tree)
at the first signof a cold, andI highly recommend it. If youcannot lo
cate it at a health food store near you, it can be obtained from (888)
406-4066 by mail order; from Rosedale Pharmacy, (888) 796-3348;
from www.rx4betterhealth.com; or at most health food stores.
Dental Infections
Quite often, dental infectionswon't be obvious, but high blood sug
ars cause dental infections, and in the typical vicious circleof diabetes,
these infectionscan causeveryhighblood sugars.You cannot possibly
control blood sugars under thesecircumstances. I have seldommet a
long-standing diabetic over age forty(witha historyof uncontrolled
blood sugars) who had all his teeth.
Frequent dental infections can be a signof diabetes for those who
have not alreadybeen diagnosed. I havehad many patients who have
undergone multiple root canal or gum treatments prior to the diag
nosis of diabetes.
If your insulin* "isn't working"—that is, your normal dose isn't
acting as you havedeterminedit should—and you havedetermined
that your insulin isn't contaminated (for example, by reusing sy
ringes) the first place to lookis in your mouth.
First,lookat your gumsto seeif there'sanysignof infection— e.g.,
redness, swelling, tenderness to pressure. Put somewaterwith crushed
icein your mouth for 30seconds. If a tooth hurts, youshouldsuspect
an infection.
Or oral agentsfor controllingbloodsugar.
Strange Biology. Phenomena Peculiar to Diabetes 101
Get anemergency appointmentwith yourdentist immediately. He
can determine if youhave asuperficial infection, andcanX-ray where
your teeth are sensitive, but he shouldrefer you to an endodontist (a
dentist whodeals withroot canals and thejawbone) oraperiodontist
(who treats infected gums). This kindof infection is extremely com
mon in diabetics and should be brought under control as rapidly as
possible in order to allow youto bring your blood sugars undercon
trol. Wewill discuss this subject further in Chapter 21.
The Laws of Small Numbers
Biginputs make big mistakes; small inputs make small mis
takes." That is the first thing my friend Kanji Ishikawa saysto
himself eachmorning on arising. It is his mantra, the single
most important thing he knows about diabetes.
Kanji is the oldest surviving type 1diabetic in Japan (he is, by the
way, younger than I, but afflicted with numerous long-term diabetic
complications because of manyyears of uncontrolled bloodsugars).
Manybiological and mechanical systems respond in a predictable
wayto small inputsbut in achaotic and considerably less predictable
wayto large inputs. Consider for amoment traffic. Put asmall num
ber of automobileson a given stretch of highwayand trafficactsin a
predictable fashion: cars can maintain speed, enter and merge into
openspaces, andexitwithaminimumof danger. There's roomfor er
ror. Doublethe number of cars andthe risksdon't just double, they in
crease geometrically. Triple or quadruple the number of cars andthe
unpredictability of asafe tripincreases exponentially.
The name of the game for the diabetic in achieving bloodsugar nor
malizationis predictability. It's very difficult to use medications safely
unless you canpredict the effect they'llhave. Nor canyou normalize
bloodsugar unless youcan predict the effects of whatyou're eating.
If youcan't accurately predict your blood sugar levels, thenyoucan't
accurately predict yourneeds for insulin or oral bloodsugar-lowering
agents. If thekindsof foods you're eating give youconsistently unpre
dictable blood sugar levels, then it will be impossible to normalize
blood sugars.
Oneof the primeintentsof thisbook isto give youthe information
youneedto learn to predict yourbloodsugar levels andhowto ensure
The LawsofSmallNumbers 103
that your predictions will be accurate. Here the Laws of Small Num
bers are exceedingly important.
PredictabiUty. Howdo you achieve it?
The oldAmerican Diabetes Association (ADA) dietary recommenda
tions allowed 150 grams of carbohydrate per meal. This, as you may
know by now, is grossly excessive for people trying to control their
bloodsugars. Hereis one reason why.
Typically, 150 grams of carbohydrate would beagood-sized bowl of
cooked pasta. You maythinkthat byreading the ingredients label on
the package you can preciselycompute howmuch ofthedrypasta you
mustweigh out to dispense exactly 150 grams of carbohydrate. Now,
if you're a nonobese type 1diabetic who weighs 150 pounds (68 kilo
grams) andmakes no insulin, 1gram of carbohydrate will raise your
blood sugar by about 5 mg/dl. By using methods that we'll later de
scribe, you cancalculate exactly howmuch insulin you mustinject to
keep your bloodsugarat thesamepoint afterthe mealasit wasbefore
the meal. This may sound elegant, but it will rarely work for a high-
carbohydrate meal. What neither theADA nor thepackage tells you is
that food producers are permitted a margin of error of plus or minus
20percent in their labeling of ingredients. Furthermore, manypack
aged products— for example, vegetable soup—cannot even match
this error range, in spite of federal labeling requirements. So even if
you perform the necessary calculations, your blood sugar after the
meal can beoffbyacarbohydrate error of5mg/dl multiplied by±30
grams (±20 percent of 150 gm), orbyawhopping ±150 mg/dl forjust
this one meal. Ifyour blood sugar level before the meal was approxi
mately 85 mg/dl, you've now got a blood glucose level anywhere be
tween 235 mg/dland 0 mg/dl. Either situation isclearly unacceptable.
Let's try another example. Say you're a type2 diabetic, obese, and
make someinsulin of your ownbut also inject insulin. You've found
that 1gram of carbohydrate only raises your blood sugar by3 mg/dl.
Your blood sugar would be off by ±90 mg/dl. If your target blood
sugar value is, say, 90 mg/dl, you're looking at a postmeal blood sugar
level of anywhere from180 mg/dlto 0 mg/dl.
That's oneof the many problems with theADA guidelines. Big in
puts and big uncertainty.
104 Before You Start
But if youeatanamount of carbohydrate that will affect yourblood
sugar by a much smaller margin of error, then you're going to havea
much simpler time of normalizing blood sugar levels. My diet plan,
whichwewill getintoin Chapters 9-11, aimsto keepthesemarginsin
the realmof ±10-20mg/dl. Howdo we accomplish this? Small inputs.
Eating only a half-teacup of pasta is not the answer. Even small
amounts of some carbohydrates cancause big swings in blood sugar.
And anyway, who would feel satisfied after such a small serving of
pasta? The keyisto eat foods thatwillaffect yourbloodsugar in avery
small way.
Small inputs, small mistakes. Sounds sosimpleandstraightforward
that it may make you want to askwhy no one has told you about it
Say thatinstead of eating pasta as thecarbohydrate portion of your
meal, youeatsalad. If youestimate 2cups of salad to total 12 grams of
carbohydrate and are off not by 20 percent but by 30 percent, that's
still an uncertaintyof only 4 grams of carbohydrate — a maximum
potential 20mg/dlrise or fall inblood sugar. Abigbowlof pasta for a
couple of cups of salad? Not much of a trade, you may say. Well, we
don't intendthat youstarve. Asyoudecrease the amount of fast-acting
carbohydrate you eat, you can often simultaneously increase the
amount of proteinyoueat. Protein can, asyou mayrecall, also cause a
blood sugar rise, but this takes place much more slowly, to a much
smaller degree, and is more easily covered with medication. In addi
tion, unlikethe pasta, whichcanleave you feeling hungry afterameal
— I will explain this further in later chapters — protein leaves you
feeling satisfiedlonger.
Intheory, youcould weigh everything youeat right downtothelast
gram and make your calculations based on information provided by
the manufacturer or derived from some of the books we use. This in
formation, asnoted above, is only anestimate, with considerable mar
gin for error. You will have onlyavague idea of what you're actually
consuming, andof the effect it willhave on bloodsugar.
The idea here is to stick with low levels of slow-acting, nutritious
carbohydrates. In addition, stickwith foods that will make you feel
satisfied without causing hugeswings in bloodsugar. Simple.
TheLawsof SmallNumbers 105
If youdo not takeinsulin, youcanskipthissection.
Thinkagain of traffic. You're driving down the road and your car
drifts slightly toward themedian. To bringit back intoline, youmake
a slight adjustment of the steering wheel. No problem. But yank the
steeringwheeland it couldcarryyou into another lane, or could send
you careening off the road.
When youinjectinsulin, not allof it reaches yourbloodstream. Re
search has shown that there's a level of uncertainty as to just how
muchabsorption of insulin actually takes place, and even as to how
sensitive the bodyis to insulin fromonedayto the next.The more in
sulin youuse, the greater thelevel of uncertainty.
Whenyouinject insulin, you're puttingbeneath your skin asubstance
that isn't, according toyour immune system's way ofseeing things, sup
posed to bethere. So a portion of it will bedestroyed asa foreign sub
stance before it can reach thebloodstream. The amount that thebody
can destroydepends onseveral factors. First ishowbig adose you inject.
The bigger the dose, themore inflammation and irritation you cause,
andthemore of a"red flag" you send uptoyour immune system. Other
factors include the depth, speed, andlocation ofyourinjection.
Your injections will naturallyvary from one time to the next. Even
the most fastidious person will unconsciously alter minor things in
the injection process from day to day. So the amount of insulin that
gets into your bloodstream is always going to have some variability.
Thebigger the dose, the bigger the variation.
Anumber of years ago, researchers at the University of Minnesota
demonstrated thatifyou inject about 20 units ofinsulin intoyourarm,
you'llget on average a 39percentvariationin the amount that makesit
into the bloodstream from one day to the next They found that ab
dominal injections hadonly a 29percent average variation, andsorec
ommended that weuseonlyabdominal injections. Onpaperthat seems
fine, but in practice theeffects onblood sugar arestill intolerable.
Sayyou do inject 20 units of human insulin at one time. Eachunit
lowers the blood sugar of a typical 150-pound adult by40 mg/dl. A
29 percent variability will create about a 6-unit discrepancy in your
20-unit injection, which means a 240 mg/dl blood sugar uncertainty
(40 mg/dlx 6 units). Theresult is totally haphazard bloodsugars and
complete unpredictabiUty, just byvirtue of the varying amounts of
insulin absorbed.
106 BeforeYou Start
Research and myown experiencedemonstrate that the smaller your
dose of insulin, the less variability you get. For type 1 adult diabetics
who are not obese, we'dideallyliketo seedosesanywhere from lA unit
to 6 units or at the most 7. Typically, you might take 3-5 units in a
shot. At these lower doses, the uncertainty of absorption approaches
zero, so that there is no need to worry about whether you should in
ject in your armor abdomen or buttock.
I havea veryobesepatient who requires 27units of long-actingin
sulin at bedtime. He's so insulin-resistant that there's no wayto keep
his blood sugar under controlwithout this massive dose. In order to
amelioratethe unpredictabilityof largedoses, he splitshis bedtime in
sulininto four smallshotsgiven into four separatesitesusingthe same
disposable syringe. Asa rule, I recommend that a single insulin injec
tion never exceed7 unitsfor adults and proportionally lessfor chil
dren, depending ontheirweight
Again, it's verydifficult to use anymedication safely unless you can
predict the effect it will have. Withinsulin, this is as true of when you
injectasit is of howmuchyoutake. Ifyou're a recent-onset type 1di
abetic, fast-acting (regular) insulin can be injected 40-45 minutes
prior to a mealtailored to yourdiet planto preventthe ensuingrisein
blood sugar. Regular, "fast-acting" insulin, despite its designation,
doesn't act very fast, and cannot come close to approximating the
phaseI insulinresponse of a nondiabetic. To a lesser degree this isalso
true of the new,faster-acting lispro (Humalog), glulisine(Apidra), or
aspart (Novolog) insulins. Still, these are the fastest we have. Small
doses of regular start to workin about 45minutes and do not finish
for at least 5 hours; lisprostarts to workin about 20 minutes and also
takes at least 5 hours to finish. This is considerably slower than the
speedat whichfast-acting carbohydrate raises bloodsugar.
Many years ago, John Galloway, then medical director and senior
scientist of Eli Lilly and Company,performed an eye-opening experi
ment. He gave one injectionof 70units of regularinsulin (a verylarge
dose) to a nondiabetic volunteer who was connected to an intra
venous glucose infusion. Dr. Galloway then measured blood sugars
everyfewminutes and adjustedthe glucose drip to keepthe patient's
bloodsugarsclampedat 90mg/dl. Howlongwouldyouguess the glu-
TheLawsof SmallNumbers 107
coseinfusion had to be continued to prevent dangerously low blood
sugars, or hypoglycemia?
It took aweek, eventhough the package insert says that regular in
sulinlasts only 4-12 hours. So the conclusion is that eventhe timing
of injected insulin is very much dependent upon how much was in
jected. In practice, larger insulininjections start working sooner, last
longer, and haveless predictable timing.
If you eat a meal not specifically tailored to our restricted-carbo
hydrate diet and try to coverit with insulin, you'll get a postprandial
(after-eating) increase in blood sugar, eventually followed by a de
crease asthe fast-acting insulin catches up. This means that you'll have
high blood sugars aftereverymeal, andyou could still fall prey to the
long-term complications of diabetes. If you try to prevent the in
evitable postprandial blood sugar spikeby waitingto eatuntil afterthe
start time of your insulin, you may easilymake yourself hypoglycemic,
which could in turn cause you to overcompensate by overeating—
that is, presuming you don't loseconsciousness first.
Type 2 diabetics have a diminished or absent phase I insulin re
sponse,and sothey face aproblemsimilar to that oftype Is. They have
to wait hours forthe phase II insulinto catchup if they eat fast-acting
carbohydrate or large amounts of slow-acting carbohydrate.
The key to timing insulininjections is to knowhowcarbohydrates
and insulin affect your blood sugar and to use that knowledge to
minimize the swings. Since you can't approximate phase I insulin
response, you have to eat foods that allow you to work within the
limits of the insulin you make or inject. If you think you'll miss out on
the ADA's great high-carb, low-fat diet — which, statistically, has only
succeededin raisinglevelsof obesity,elevatingtriglycerides and LDL,
and causingan epidemic of diabetes and earlydeath— there is con
siderable evidencethat restricting carbohydrate is healthier not only
for diabetics but for everyone. This has recently been supported by
a twenty-year study of 82,802 nondiabetic nurses published in the
November 9,2006, issueof the NewEngland Journal ofMedicine. (For
more details on this point, see Protein Power, by Drs. Michael and
Mary Dan Eades, BantamBooks, 1996.)
If you consumeonly small amounts of slow-acting carbohydrate, you
can actually prevent postprandial blood sugar elevation with injected
preprandial rapid-acting insulin. In fact, by restricting carbohydrate in
take, many type 2 diabetics will be able to prevent this rise with their
phase IIinsulin response andwill not need injected insulinbefore meals.
108 BeforeYou Start
Essential to obeyingthe Laws of Small Numbers is to eat only small
amounts of slow-acting carbohydratewhenyou eat carbohydrate, and
no fast-acting carbohydrate. Even the slowest-acting carbohydratecan
outpace injectedor phase II insulin if consumed in greater amounts
than recommended later in this book (Chapters 9-11).
If you eat a small amount of slow-acting carbohydrate, you might
get by with a very small or no postprandial blood sugar increase. If
youdoublethe amount of slow-acting carbohydrate, you'llmore than
doublethe potential increase inbloodsugar(andremember that high
blood sugar leads to even higherblood sugar). If youfill up on slow-
acting carbohydrate, it will work asfast asalesser amount of fast-acting
carbohydrate, and if youfeel stuffed, you'llcompound it with the Chi
nese restaurant effect.
All of this not only points toward eatingless carbohydrate, it also
implies eatingsmaller meals 4 or 5timesa dayrather than threelarge
meals. Ifyou're a type2diabetic andrequire no medication, eatinglike
this mayworkwell for you. The difficulty with this sort of plan is its
inconvenience, but somepeople don't mind and actually preferto eat
this way. One of my patients, a type 1 diabetic who still makes some
insulin, eats a couple of bites of protein every 20 minutes and takes
long-acting insulin. In a 16-hour day, that adds up to a lot of mini-
meals and a lot of clock-watching. This routine would drive many
peoplenuts, but it almostworks for her.As longas she keeps up with
her frequent little meals and covers the insulin, she's fine. When she
misses a few"meals," there inevitably is trouble.
Forthe type2diabetic who doesn't needinsulin injections, smaller
meals throughout theday can beavery effective way of maintaining a
constant level of blood sugar. Sincethis kind of diet would be tailored
to work with a phase II insulin response, blood sugars should never
gotoohigh. It would, however, involve acertain amountof dailyprep
aration and routinization that could be thrown off by changes in
schedule —illness, travel, houseguests, and so forth. People who
cover their meals with injected insulin and also correct small blood
sugar elevations with very rapid acting insulin, however, cannot get
awaywith more than three dailymeals(Chapter 19).
Establishing a Treatment Plan
Nowthat you know the different factors that can affect blood
sugar, we can begin to discuss treatment plans. Blood sugar
normalization for most diabetics canbe achieved throughone
of four basic plans. Although there are only two major types of dia
betes— type 1andtype 2 — there are somanyvariations, particularly
in type 2, that atreatment planthat works for one diabetic won't neces
sarily work for another. Each planhasto be tailored to the individual.
The basictreatment plans increase in complexity with the severity
of the disease.
For type 2 diabetes
Level 1: Diet (and appropriateweight loss)*
Level 2: Diet (and appropriateweight loss) plus exercise
Level 3: Diet (and appropriate weight loss) plus exercise plus an
oral insulin-sensitizing or insulin-mimetic agent
Level 4: Diet (and appropriate weight loss) plus exercise plus in
sulin injections, with or without an oral agent
Fortype 1 diabetes
Same aslevel 4 above, with the addition of multiple daily insulin
injections, with questionablebenefit from exercise in controlling
blood sugars, andwith benefit fromoralinsulin-sensitizing agents
* Since 80 percent or more of type 2 diabeticsareoverweight, weight loss should
be an important part of treatment for the majority.
110 Before You Start
only when insulin requirements are excessive, aswith those who
are obese or who have polycystic ovariansyndrome (PCOS; see
Appendix E).
What are normal blood sugar levels? What range do we findin nondi
abetics? The answers depend upon whom you ask. I've seen figures in
the scientificliteratureoverthe years ranginganywhere from 60 to 140
mg/dl. My experience checking random blood sugar readings on
nonobese nondiabetics, aswell as figures fromlarge population stud
ies, tells me that for most nondiabetics, blood sugar levels cover a
prettynarrowrange of about75-95mg/dl (by finger stick), except af
ter mealscontaining large amounts of fast-acting carbohydrates.
I usually select a target of 90 mg/dl for most of my patients who
take insulin. This target is not an average, but one we try to maintain
24hours aday. Evenif you average 90mg/dl but your blood sugars are
bouncing backand forthbetween60and 140 mg/dl, you'restill on the
roller coaster. Our object is to find a treatment planthat will get you
off the roller coaster and keep you off.
For those who do not need insulin injections to maintain blood
sugars, and those insulin users who have demonstrated very stable
blood sugars, I eventuallyset a target of 80-85 mg/dl. This assumes
that you're comfortable at such levels, that is, not experiencing symp
toms of hypoglycemia (lowblood sugars).
One of the most important considerations in setting up an initial
target is that people who have had high blood sugar levels for many
months or years usually experience unpleasant symptoms of hypo
glycemia asblood sugars approach normal. Someone who has grown
accustomed to blood sugars consistently over 300 mg/dl may feel
"shaky"at 100 mg/dl. In suchacase, we might start with 160 mg/dl as
the initial target.We'd then lowerthe targetto its ultimate value over a
periodof weeks or months astreatment proceeds.
It's unusual when aninitial meal planand dosage of medication in
stantly result in the desired blood sugar profiles. Some people, a few
days into their regimen, may find something objectionable, such as
not enough to eat for a certain meal. Because of this, it's often neces
saryto experiment with aplan, makingsmallchanges basedupon per
sonal preferences and blood sugar profiles.
Establishinga Treatment Plan 111
People tend to become discouraged if they cannot see rapid im
provement, and so, where warranted, I try to make adjustments to the
regimen everyfewdaysin order to demonstrate that our efforts are ac
complishingpositiveresults. Tothis end, I ask patients to bring or to
faxto myofficetheir blood sugar profilesabout one weekafter their fi
nal training visit, if initial treatment is by diet alone. If I've prescribed
insulin, I like to see profiles within a few days. I certainlytry to make
sure that no blood sugars are below70mg/dl during this trial period.
I ask all new patients to phone me at any time of the day or night if
they experience a blood sugar under 70 or become confused about
their instructions. Additional repeat visitsor phone callsmay be nec
essary every fewdays or weeks, depending upon how rapidly blood
sugar profilesreach our ultimate target.
Many newpatients come to my officefrom out of town, some trav
eling distances of thousands of miles. Clearly, frequent office visits
wouldbe impractical in suchcases. For thesepatients, I oftenschedule
follow-up "telephone visits" instead of office visits. Patients will fax
their blood sugars to me on Glucograf III data sheets.
These subsequent office or telephone interactions enable me to
fine-tune the original plan, and alsoto reinforce the training program
by catchingany mistakesthat a patient mayinadvertentlymake. This
interactivetraining is much more effective for patients than just read
ing a book or hearing a fewlectures.*
Although the protocol will likelydiffer at everydoctor's office, in the
next several pages, I'll try to give youan ideaof howthings work at our
Diabetes Center. This way, you'll get a general notion of how a com
prehensive diabetes treatment program should work.
In my experience, most patients will cooperate with a treatment
planthat shows themconcrete results. Greatly improved bloodsugars,
weight normalization, halting or reversing diabetic complications,
*Nevertheless, I record my 4-6 hour training sessions for my patients and give
themthe tapes. Readers of this book canpurchaseCDrecordingsof actual train
ing sessionsat www.rx4betterhealth.com or (800) 798-6922.
112 BeforeYou Start
and a senseof improvedoverall health can goa long waytoward con
vincingan individual to stickwitha treatment program.
Much is written in the diabetes literature about the key role of pa
tient "compliance." Treatmentfailures are oftenblamed upon "lack of
compliance." I think it's unreasonable to expect anyone to comply
with a treatment plan that explainslittleand, as in the caseof the stan
dard ADA approach, isn't really effective and offerslittle incentive to
continue. What wemust do isset up a sensible,workableplan that you
understand and agree with. When I work with my patients in the
office, I don't just havemystaffhand thema photocopieddiet and ex
pect automatic acceptance. This is something that has to be negoti
ated, worked out. Do you liketurnips? Great, wecan probably fit them
into your diet, but I don't think I've ever eaten one in my life. Call it
"physician compliance," but the point isthat it's unreasonableto try to
force mypersonalpreferences on mypatients. Onlywhenone under
stands and agrees with the plan canweexpectcooperation. For coop
eration to continue, however, patients haveto seepositive, rapid results.
Not all people are able to follow a given treatment plan. For exam
ple, someone who's been overeating carbohydrate for a lifetime may
find it next to impossibleto begin to followa restricted diet immedi
ately,but we have waysaround this (see Chapter 13). Some absolutely
resist exercise. But for most people we are still able to develop a treat
ment plan that works. If, for example, someone whose blood sugar
should be controllable with diet and exercise refuses to exercise, I will
instead prescribe medication that lowersinsulin resistance.
When seeingnewpatients, for those who livenearby, my preference is
an introductory visit followed later by a series of treatment/training
visits lasting 2-3 hours each. The continuity of time is invaluable to
showingrapid results. However, most insurance companies don't like
to pay for lengthy office visits— especially for diabetes training —
and so it may be necessaryto break down the initial workup and train
ing into multiple brief visits. Although I don't like to, I may do this
with local patients; but with patients who livea great distance from my
office, it's simply not workable to have successive short visits.
At the first visit I always get a drop of fingertip blood to measure the
patient's baseline (initial) HgbA,c. As time goes on and the patient
Establishinga Treatment Plan 113
sticks with the program, the inevitable progression of reduced blood
sugar over the next few months can providetremendous encourage
My preferred procedure for the first few days of treatment is to
break down visits into three sessions.
Introductory Visit
Since bloodglucose profiles aresoessential to formulatinga treatment
plan, prior to the introductory visit I usuallyask a newpatient to pro
cure blood glucosetesting supplies — Glucograf III data sheets and
the other supplies listed in Chapter 3.1 provide guidelines for blood
glucose self-monitoring (like those you haveseen in Chapter 4), and
askthe patient to learn howto use the equipment so that later, on the
first treatment/training visit, I can look over one or two weeks' blood
glucose profiles. I alsomaygive the patient a coupleof largebottlesso
that a 24-hour urine specimencan be collected for a subsequent visit.
First Treatment/Training Visit
If I haven't done so in the introductory visit, I take a medical history
and begin a physical examgearedtowarduncoveringlong-term com
pUcations of diabetes. For patients who havehad diabetes more than
about five years, I inevitably find a good number of these long-term
sequelae (aftereffects), some of whichmaybe reversed byblood sugar
normalization. The exam will include tests described in Chapter 2. We
check to ensure the patient has purchased the right supplies. If we
haven't done so already, we provide a supply list (Chapter 3) with ap
propriate items checked off.
We discuss plans for treatment of medical problems other than
blood glucose control. These may include conditions the patient al
readyknows about, but also anythinguncovered by blood testing or
by the physical exam. If the patient has already acquiredsuppliesand
begun measuring blood sugars, I reviewhis or her technique and cor
rect it if necessary.
Second Treatment/Training Visit
Manyof my patients come from out of town, and so the second visit
maytakeplacethe dayafter the first. For localpatients,however, it will
be approximately a week later. At this visit we finish the physical ex
amination. We also recheck the patient's blood glucose measurement
technique and his use of the Glucograf form.
114 Before You Start
If I feel that the patient should be taking insulin, I give instructions
for insulin doses to be takenthe night before and the morning of the
third visit. I also providetraining in self-injection (Chapter 16) to pa
tients who haveneverinjectedbefore.For those who areveteraninsulin
users, I evaluate self-injection techniqueandcorrect it if necessary. It's
my experience that most insulin-using patients have previously been
taught improper techniques for filling syringes and injectinginsulin.
To this visit the patient is expectedto bring the blood sugardatahe
or she has collectedoverthe priorweek(s), together with aseparatelist
of what he/she eats on a typical day. This information enables me to
estimateif the patientwill needmedication for blood glucose control
and tells me about foods the patient likes that might be included in
our meal plans. The blood glucose profile also providesa snapshot of
the patient's status before beginning the new treatment regimen. We
can reviewthis at alaterdateto evaluate progress. As with eachof the
other initialvisits,the bulk of our time will be devotedto training.*
Most important, this is the visit where we negotiate the meal plan
(see Chapter 11).
Third Treatment/Training Visit
This visit may take place anytime afterthe second.We ask the patient
to come in fastingand to bring a 24-hour urine collection. At this visit
I drawblood for baselinestudies and continue training. I alsoenter all
the "datato remember"at the top of aGlucograf datasheet (Chap
ter 5). I also use this visit to give verbal instructions and a printed
handout regarding foot care (seeAppendix D).
Patients to be treated with insulin may be kept fasting until supper
on the day of this visit in orderto determine if the small basal dose of
long-acting insulin that was injected that morning is adequate to
maintain blood glucose at a fixed level. On this day, ifthe patient arises
with ablood glucose above our target value, she'd have instructions to
take atrial dose of fast-acting insulin to bringblood sugardown to the
target value. If blood sugar on awakening isbelowthe target, she'duse
glucose tablets to bringblood glucose up to target. By this means, we
confirm or correct my estimation of how much a givenamount of in
sulin or glucosewill loweror raise the individual'sblood sugar.
* My training program consists essentiallyof the material covered in this book.
It's my hope that physicianswho havelittle time to educate patients will use this
book to assist in that purpose.
Establishinga Treatment Plan 115
Whenever I talkabout bloodsugars inthisbook, I'mreferringto finger-
stick, plasma blood glucose measurements. When I discuss "normal"
blood sugar values, I am referringto those found in nonobese non
diabetics — and to those not taken within 3 hours of a high-
carbohydrate meal.
In my experience, given the right blood sugar meter, these values
will be almost exactly the same as you wouldget from plasma mea
surements of venous blood that your doctor would send to a clinical
laboratory. I've seen finger-stick blood sugars measured on hundreds
of nondiabetic, nonobese adults (for example, salespeople who come
into the office trying to sellme meters — I insist on demonstrations;*
or the nondiabetic spouses, parents, or siblings of patients). It usually
isabout 83mg/dl.In order to simplify, I round offand tellmypatients
that a normal to shoot for is 85 mg/dl, no matter what age. I haven't
had the opportunity to test a great number of nondiabetic children,
but the literatureshows that normalbloodsugars willbeabout 85mg/
dl, with the potential to be considerably lower.*
With respect to hemoglobin A,c, I have a sophisticated machine in
myoffice that I'vefound correlates almostexactlywith measuresfrom
aclinical laboratory. I therefore check HgbAlc values onevery patient
at everyroutine visit, and frequently on nondiabetic relatives. Essen
tially what I see isthat nondiabetics who are not obese have HgbAlc
levels in the rangeof 4.2-4.6percent. I have a number of diabetic pa
tients who, under treatment, nowhave HgbAlc readings aslow as4.2
*I usedtohavesomefunwithnondiabetic sales repswhentheycameintotheof
fice selling bloodsugarmeters. They'd bedemonstrating a meter, which I would
compare to my own meter. I always used their bloodbecause I've had enough
finger sticks. I'd "guess" their blood sugar. I'd make a show of examining their
skin, then give them a number. It was always about the same, but they didn't
knowthat. The number was 83 mg/dl. Inevitably I'd be within ±3 mg/dl. You
know, of course, that I didn't have anyspecial powers —it wasjust that I'd seen
so many random finger-stick readings from nondiabetics, I knewwhat number
the nondiabeticwaslikelyto show.
t Astudypublishedin the New EnglandJournal ofMedicine found that nondia
beticmenwith fasting bloodsugars of 87mg/dl or morehad progressively in
creased riskof developing diabetes than those with values less than 81 mg/dl.
Another study of about 2,000healthy men, published in Diabetes Care in Janu
ary 1999, showedthat over a period of twenty-two yearsthe riskof cardiacdeath
was40percent greaterfor thosewithfasting bloodsugarsgreaterthan 85mg/dl.
116 BeforeYou Start
percent. This is a considerable deviation from the ADA's recommen
dation of under 6 percent — with no intervention unless levelsexceed
7 percent. In my opinion, this is yet another example of "the rape of
the diabetic."
The ADA recommendation for "tight control" of blood sugars,
from its Web site, is as follows:
Ideally,this means levels between 90 and 130mg/dl before meals
and less than 180two hours after starting a meal, with a glycated
hemoglobin level lessthan 7 percent.
The recommendations go on to state that tight control (what I advo
cate) "isn't for everyone," which I believe is nonsense. But the ADA's
tight control as definedabove isn't verytight at all.I would call it "out
of control."
Many years ago, I reviewed dozens ofHgbAlc values andthousands of
blood sugars from data sheets submitted by my patients and came up
witha formula forconverting HgbAlc to mean (average) bloodsugar.
Myformuladoesnot jibewithmost other formulas, perhapsbecause
others haven't collectedblood sugars throughout the day running into
the hundreds or eventhousands ofpatients covering4-month periods.
Theformula isverysimple. An HgbA,c of 5percent isequivalent to an
average blood sugar reading of 100 mg/dl,and every1percent above5
corresponds to an additional 40 mg/dl increase in blood sugars. Soan
HgbAlc of 7 percent would correspond to an average blood sugar of
Theformula is, in myexperience, useless for HgbAlc values of less
than 5 percent, and it may not work for average blood sugars greater
than 300 mg/dl for the simple reason that for a new patient running
blood sugars greater than 300 mg/dl, we rapidly get them down into
the 100s or less. Such new patients don't come in bringing me hun
dreds of data points in the 300s for me to compute an accurate for
mula at these values— nor would I askthem to. Many may not bring
me any prior blood sugar data on their initial visit.
In February 2002a study publishedin Diabetes Care reported a for
mula that is valid for average blood sugars over a much wider range
Establishinga Treatment Plan 117
than mine, includingvalues well above andbelow100 mg/dl. It gives
results close to mine in the 100-200 mg/dl range. The formula is:
mean plasma glucose =(35.6 x HgbAlc) - 77.3.
So how do we goabout settingatarget normal value givenallthese
numbers? Let's take a look at a type 2 diabetic whose disease can be
controlled by diet and exercise. Here, we'll certainly shoot for blood
sugars of about 83 mg/dl before, during, and after meals. It will then
be up to both me and the patientjointly— if hisblood sugars are, say,
in the 90s — to decide whether we want to introduce medications to
further lower blood sugar. Many patients these days are hesitant to
take any medicationthat's been approved by the FDA, despite many
suchmedications' beingquitebenign. If wehave atype 2 diabetic who
requires the insulin-sensitizing drugs like metformin or the thiazoli-
dinediones, we certainly canshoot for atarget blood sugar of 83mg/dl
before, during, and after meals, and indeed, I will work with the pa
tient to juggle the medications,usinglong- or short-actingversionsin
orderto achieve that target.
Type 2 diabetics who require very small amounts of insulin (say,
1-2 units perdose) are at verylowrisk for hypoglycemia andwill usu
allyautomatically"turn off" the insulinthey makethemselvesif blood
sugars are too low. Suchpeopleare also goodcandidates foratarget of
83 mg/dl.
When it comes to type 1diabetics, wherevirtuallyallof the needed
insulin is going to be injected, I increase the target to 90 mg/dl, even
though we know that the mortalityrate — evenin the general, nondi
abetic population— is slightly greater for those with fasting or post
prandial blood sugars of 90 mg/dT than it is for those with blood
sugars of 83. If at all feasible without frequent hypoglycemic episodes,
I will eventuallylower the target to 83 mg/dl. I now use 83 as a target
for myself.
A target may imply corrections to get you to your target. As a rule,
if you're a type 2, your blood sugar goes down eventually— maybe
quickly, maybe over many hours. If you'rea type 1 and injecting sig
nificant dosesof insulin, if you make amistakeon your diet and your
blood sugargoes up, you haveto inject additional, calibrated doses of
fast-acting insulin deliberately to bringdown your blood sugar and, if
it's too low, take glucosetablets to raise it.
For a new patient in the very early stages of type 2 diabetes, I may
seeboth hypo- andhyperglycemia. This is probablybecauseone ofthe
early"lesions" of type 2 is difficulty in storing the insulin granules
118 BeforeYou Start
your body makes. So such a person would make insulin for a meal,
then make more after the meal. A nondiabetic would store that addi
tional insulin as it's being made, but the early type 2 would release
some or all of it into the bloodstreamas it's generated, thereby bring
ing blood sugar too low. This explanationalsoaccountsfor attenuated
(diminished) phase I insulin response—just not having enough in
sulin stored to cover a meal adequately (another reason to follow a
low-carbohydrate diet). Such an individual could experience blood
sugars in the 70s or evenmid-60s from time to time, and these indi
vidualsmust carryglucose tabletswith themto bring blood sugarsup
to their target, usually 83. They don't take injected insulin to bring
blood sugar down if it goes too high when they make a mistake, be
causetheir bodies will do that for them, probablyfaster than injected
insulin would.
On the third visit, it's generally appropriate to prepare a fist of treat
ment goals. Exactly what are we going to accomplish, how, and over
what time frame? The patient and I discuss a list of goals to make sure
that he or she understands and agrees. The following list is typical of
the things I want to see any givenpatient accomplish. (Remember, the
training I provide to my patients is the substance of this book, so if
you don't entirelyunderstand all of thesegoals right now,don't be dis
couraged. Mark this chapter and comebackto it whenyou'vefinished
the book. By then you should understand the whole philosophy of
my approach and the goals will make sense. You may also by that
time have developed — if you haven't already— conscious goals of
your own.)
• Normalization of blood glucose profiles.
• Improvement or normalization of the following laboratory tests
that respond to blood glucosecontrol (Chapter 2):
hemoglobin A1C thrombotic riskprofile
red blood cell magnesium renal profile
lipid profile
• Attainment of ideal weight (whereappropriate).
• Full or partial reversal of diabeticcomplications, including pain
or numbness in feet, diabetes-related retinal or kidneyproblems,
Establishinga Treatment Plan 119
gastroparesis, cardiac autonomicneuropathy, neuropathic erec
tiledysfunction, postural hypotension, andsoon. If bloodsugars
are kept normal, someof these improvements will appear within
weeks to years, depending upon the particular problem and its
• Reduction in frequency and severity of hypoglycemic episodes
(where appropriate).
• Relief of chronic fatigue and short-term memory impairment
associated with high blood sugars.
• Improvement or normalization of hypertension.
• Reductionof demandupon betacells. If C-peptideis present be
fore starting our program (that is, if the pancreas is producing
measurable amounts of insulin), glucose tolerance should im
prove if a regimen is pursued that minimizes the demand upon
the beta cells. This is a very important goal. Remember that for
type 2 patients,small sacrifices nowcanpreventthe need for 5 or
more daily insulin doses down the road. Beta cell burnout (see
page99) can frequently be prevented.
• Increased strength, endurance, and feeling of well-being.
The patient may wish to add some personal goals. The doctor
should respect these if at all possible. For example, I have several pa
tients who arewillingto dowhateverI ask,providedI do not put them
on insulin. I consider this a reasonable preliminary goal for some,
even though it may increase the risk of beta cell burnout. After all, if
we cannot enlist a patient'scooperation,we achievenothing.
The Basic Food Groups
In Chapter 1 we discussed how diabetics and nondiabetics might
reactto a particularmeal. Herewe'll talk about howspecifickinds
of foods canaffect your blood sugar.
A curious fact about diet, nutrition, and medication is that while we
canmake accurate generalizations about how most of us will react to
a particular diet or medical regimen, we cannot predict exactly how
eachindividual will reactto a given food or medication.
The foods we consume, once you take away the water and undi-
gestible contents, canbe groupedinto threemajor categories that pro
vide calories or energy: protein, carbohydrate, and fat. (Alcohol also
provides calories, and will be discussed laterin this chapter.) Seldom
will food fromone of these groups containsolelyone type of nutrient.
Proteinfoods often contain fat; carbohydrate foods frequently contain
some proteinand some fat. The common foods that are virtually 100
percent fat areoils, butter, some types of margarine, and lard.
Since our principal concern here is blood sugarcontrol, we'll con
centrate on how these three major sources of calories affect blood
sugar. If you're a long-standing diabetic and have followed standard
ADAteachings for years, you'll findthat much of what you'reabout to
read is radically at odds with the ADA's dietary guidelines — andwith
good reason, asyou'll soon learn.
When we eat, the digestive process breaks down the three major
foodgroupsinto their buildingblocks. Thesebuildingblocks are then
absorbed into the bloodstream and reassembled into the various
products our bodies need in orderto function.
124 Treatment
Proteins are constructed of building blocks called amino acids.
Throughdigestion, dietary proteins are broken downby enzymes in
the digestive tract into their amino acid components. These amino
acids can then be reassembled not onlyinto muscle, nerves, and vital
organs, but also into hormones, enzymes, and neurochemicals. They
can alsobe convertedto glucose, but veryslowly and inefficiently.
Weacquiredietaryproteinfroma number of sources, but the foods
that are richest in it —eggwhites, cheese, and meats (including fish
and fowl) — contain virtually no carbohydrate. Protein is availablein
smaller amounts from vegetable sources such as legumes (beans),
seeds, and nuts, which also contain fat and carbohydrate.*
Protein and carbohydrate are our two dietary sources of blood
sugar. Protein foods from animal sources are only about 20 percent
protein by weight (about 6 grams per ounce), the rest being fat, water,
and/or undigestible "gristle." The liver (and to a lesserdegree,the kid
neys and intestines), instructed by the hormone glucagon,* can very
slowlytransform as much as 36 percent of these 6 grams per ounce
into glucose* —if bloodsugardescends too low, if seruminsulinlev
els are inadequate, or if the body's other amino acid needs have been
met. Neither carbohydrate nor fat can be transformed into protein.
In many respects— and going against the grain of a number of
the medical establishment's accepted notions about diabeticsand pro
tein — protein will become the most important part of your diet if
you are going to control blood sugars, just as it was for our hunter-
gatherer ancestors.
If you are a long-standingdiabeticand are frustrated with the care
you've received over the years, you haveprobablybeen conditioned to
* Phosphate, a by-product of protein digestion, requires calciumin order to be
eliminated fromthebody—about 1gramof calciumfor every 10ouncesof pro
tein foods. If youdon't eat muchcheese, cream,milk (too high in carbohydrate),
yogurt, or bones, all good sources of calcium, it would be wise to take a calcium
supplement. This will prevent slowloss of calcium from your bones. I recom
mend calciumin formulations supplementedwith magnesiumand vitamin D.
+Andother so-called counterregulatory hormones, suchas Cortisol and growth
$Thisamounts to about 7.5percentof the total weight of a proteinfood.Sayyou
eat a 3-ounce (85 grams) hamburger, no bun, for lunch — the protein in it can
slowlybe transformed by the liverinto no more than 6 grams of glucose.
TheBasicFood Groups 125
think that protein is more of a poisonthan sugar and is the causeof
kidneydisease. I wasconditionedthe sameway— manyyears ago, as
I mentioned, I had laboratory evidence of advanced proteinuria, sig
nifyingpotentially fatal kidneydisease — but in this case,the conven
tional wisdomis just a myth.
Nondiabetics whoeat a lot of proteindon't get diabetic kidneydis
ease. Diabetics with normal blood sugars don't get diabetic kidney
disease. High levels of dietaryprotein do not cause kidneydisease in
diabetics or anyone else. There is no higher incidence of kidneydis
ease in the cattle-growing states of the United States, where many
people eat beefat virtually every meal, than thereisin the stateswhere
beef is more expensive and consumed to a much lesser degree. Simi
larly, the incidence of kidney disease in vegetarians is the sameas the
incidence of kidneydisease in nonvegetarians. Itisthe high bloodsugar
levels that are unique to diabetes, andto a much lesser degree the high
levels ofinsulin required tocover high carbohydrate consumption (caus
inghypertension), that cause the complications associated withdiabetes.
The Big Fat Lie
Callit the BigFat Lie. Fat has, through no realfault of its own, become
the great demon of the American dietary scene. It is no myth that
more than half of Americans areoverweight, and the number of obese
Americans is growing.
Current dietary recommendations from the government, and
nearly every "reputable" organization with an opinion, are to eat no
more than 35 percent of calories as fat —which veryfew peoplecan
maintain—and therearesomerecommendations for evenlowerper
centages than that. The low-fat mania in our culture has spawnedan
increase in sugar intake. All a candy or cookie has needed is the label
"fat free" to sendits sales throughthe roof. The fallacy that eatingfat
will make you fat is about as scientifically logical as saying that eating
tomatoes will turn you red.
This is the kind of fallacious thinking behind the prevailing "wis
dom," which maintains that there is an unavoidable link between di
etary fat and high serum cholesterol. And that if you want to lose
weight and reduce cholesterol, all you need to do is eat lots of carbo
hydrate, limit consumption of meat, and cut out fat as much as possi-
126 Treatment
ble. But many contemporary researchers exploring this phenomenon
havebegun to arrive at the conclusion that a high-carbohydrate diet,
especially rich in fruit and grain products,is not so benign. In fact, it
has been shown — and it is my own observationin myself and in my
patients— that such a diet can increase body weight, increase blood
insulin levels, and raise most cardiac risk factors.
In an unbiased, clearheaded, and award-winning article in the re
spected journal Science of March 30, 2001, the science writer Gary
Taubes explores what he calls "The Soft Science of Dietary Fat." (A link
to the full text of this article is available at www.diabetes-book.com.)
Taubes cites the failure of the antifat crusadeto improve the health of
Since the early 1970s, for instance, Americans' average fat in
take has dropped fromover40%of total calories to 34%;average
serum cholesterol levelshave dropped aswell
Meanwhile, obesity in America, which remained constant
from the early 1960s through 1980, has surged upward since
then — from 14%of the population to over 22%. Diabetes has
increased apace. Both obesity and diabetes increase heart disease
risk, which could explainwhy heart disease incidence is not de
creasing. That this obesityepidemic occurred just asthe govern
ment began bombarding Americans with the low-fat message
suggests the possibility... that low-fat diets might have unin
tended consequences — among them, weight gain. "Most of us
would have predictedthat if we canget the population to change
its fat intake, with its dense calories,* we would see a reduction in
weight," admits [Bill] Harlan [of the NIH]. "Instead, we seethe
exact opposite."
I urge you to have alook at the article, which will giveyou a notion
of the kinds of competing personal, economic, and political interests
that gointo the formulation of"scientific" guidelines.
The U.S. Centers for Disease Control and Prevention (CDC) re
leased datain the year 2004 indicatingthat 66.4percentofU.S. adults
were overweight and 32.2 percent were obese. Furthermore, the inci
dence of overweight in children and adolescents aged 2-19 years in-
*Contrary to traditional thinking,a studyrecently publishedin the Journal ofthe
American College of Nutrition demonstrated that the metabolizable calories in
fats are about the same as in carbohydrates.
The Basic Food Groups
19JS 1902 1969 1976
Fig 9-1. From 1955to 1990, even as the
percentage of calories consumed asfat
declined, thepercentage ofoverweight
Americans increased by nearly half.
1983 1990
creased from 11 percent to 19 percent in the period 1988-94 through
2003-04. These statisticsareoccurringeven though people areeating
less fat.
The advent of our agricultural society is comparatively recent in
evolutionary terms — that is, it began only about 10,000 years ago.
For the millions of years that preceded the constant availability of
grain andthe more recent year-round availability of avarietyof fruits
and vegetables, our ancestors werehunters and atewhat was available
to them in the immediate environment, primarily meat, fish, some
fowl, reptiles, and insects — food that was present year-round, and
predominantly protein and fat In warm weather, some may have
eaten fruits, nuts, and berries that were available locally in some re
gions and not deliberately bred for sweetness (agriculture didn't ex
ist). If they stored fat in their bodies during warm periods, much of
that fat was burned up during the winter. Although for the past two
centuries, fruit, grain, and vegetables have, in one form or another,
been available to us in this country year-round, our collective food
supply has historically been interrupted often by famine— in some
cultures more than others. The history of the planet as best as we can
determineis one of feast (rarely) and famine, and suggests that famine
will strike again and again asit has in the last fewdecades in avariety
of places.
128 Treatment
Curiously, what todayseems in our society to be a genetic predis
position toward obesity functioned duringthe famines of prehistory
as an effective method of survival. Ironically, the ancestors of those
whotodayaremost at riskfor type2diabetes were, duringprehistory,
not the sickand dying, but the survivors. If famine strucktodayin the
UnitedStates, guess whowould survive most easily? The samepeople
who are most at riskfor type 2 diabetes. For those livingin a harsh en
vironment wherethe availabilityof foodis uncertain, bodies that store
fat most efficiently when food is available (for example, by being
insulin-resistant and cravingcarbohydrate, likemost type 2 diabetics)
survive to reproduce.
If you give it somethought, it makes perfectsense: If a farmer wants
to fatten up his pigs or cows, he doesn't feedthem meat or butter and
eggs, he feeds them grain. If you want to fatten yourself up, just start
loading up on bread, pasta, potatoes, cake, and cookies — all high-
carbohydrate foods. If you want to hasten the fattening process, con
sume dietary fat with your carbohydrate. Indeed, two recent studies
showed that dietaryfat, whenconsumed aspart of a high-carbohydrate
diet, was converted to body fat. Fat consumed as part of a low-carbo
hydrate diet was metabolized,or burned off.
The Insulin/Fat Connection
The primary sourceof bodyfat for most Americans is not dietary fat
but carbohydrate, which is convertedto blood sugar and then, with the
aid of insulin, to fat by fat cells. Remember, insulin is our main fat-
building hormone. Eat a plate of pasta. Your blood sugar will rise and
your insulin level (if you havetype 2 diabetes or are not diabetic) will
alsorise in order to cover, or prevent, the jump in blood sugar. Allthe
bloodsugarthat isnot burnedasenergyor storedasglycogen isturned
into fat. Soyou could, in theory, acquire morebodyfat fromeatinga
high-carbohydrate "fat-free" dessert than youwouldfromeatinga ten
der steaknicely marbledwithfat. Even the fat in the steakismore likely
to be stored if it is accompanied bybread, potatoes, corn, and so on.
The fatty-acid building blocks of fats can be metabolized (burned),
stored, or converted byyour bodyinto other compounds, depending
on what it requires. Consequently, fat is always in fluxin the body,be
ing stored, appearing in the blood, and being converted to energy.The
amount of triglycerides (the storage form of fat) in your bloodstream
at anygiventime willbe determinedbyyour heredity, your levelof ex
ercise, your blood sugar levels, your diet, your ratio of visceral (ab-
TheBasicFoodGroups 129
dominal) fat to lean body mass (muscle), andespecially byyourrecent
consumption ofcarbohydrate. The slimand fit tend to bevery sensitive
(i.e., responsive) to insulin and have low serum levels not only of
triglycerides but insulinaswell. But eventheir triglyceride levels will
increase after ahigh-carbohydrate meal, as excess bloodsugar is con
verted to fat. The higher the ratioof abdominal fat (and, to alesserde
gree, total body fat) to lean body mass, the less sensitive to insulin
you'll tendto be. In the obese, triglycerides tendto be present at high
levels in thebloodstream all thetime. (This is sometimes exaggerated
duringweight loss because fat is appearing in the bloodstream as it
comes out of storage to be converted into energy.) Not only are high
triglyceride levels a direct cause of insulin resistance, but they also
contribute to fatty deposits on the walls of yourbloodvessels (athero
sclerosis). Research demonstrates that if high concentrations of
triglycerides are injectedinto the blood supply of the liver of a well-
conditioned athlete, someone very sensitive to insulin, she will be
come temporarily insulin-resistant. (The most important thing to
note here is that insulin resistance, as well as other risk factors for dia
beticcomplications, canbe reversed by eating less carbohydrate, nor
malizing blood sugars, and slimming down, which we'll discuss in
greaterdetail later on.)
If you become overweight, you'll produce more insulin, become
insulin-resistant (which will require youto produce yetmoreinsulin),
andbecomeevenmore overweight because you'llcreate more fat and
store more fat. You'll enterthe vicious circle depicted in Figure 1-1.
Consider that steak I mentioned earlier. Asyouknow,the body can
convert proteinto blood sugar, but it does so at avery slowrate, and
inefficiently. Seruminsulinlevels derived from the phase II insulinre
sponse or even from insulin injectedbefore a meal may thus be suffi
cient to preventablood sugar rise fromproteinconsumption by itself.
Dietary fat cannot be converted to blood sugar, and therefore it
doesn't cause seruminsulinlevels or requirements for injected insulin
to increase. Say you eat a 6-ounce steak with no carbohydrate side
dish — this won't require much insulin to keep your blood sugar
steady, and the lower insulin level will cause only a small amount of
the fat to be stored.
Now consider what would happen if you instead ate a "fat-free"
dessertwith exactlythe samenumber of calories asthat steak.Your in
sulin level will jump dramatically in order to cover the sugar and
starches in the dessert. Remember, insulin is the fat-building and fat-
130 Treatment
storage hormone. Since it's dessert, youprobablywon'tbe going out to
run amarathonafter eating, sothe largest portionof your newlycre
atedblood sugar won't getburned. Instead much of it will be turned
into fat and stored.
Interestingly enough,eating fat with carbohydrate canactually slow
the digestion of carbohydrate, so the jump in your blood sugar level
might thereby be slowed. This wouldprobably be relatively effective if
you're talking about eating a green salad with vinegar-and-oil dress
ing. But if you're eating aregular dessert, or abaked potato with your
steak, the slowdownin digestion would not prevent blood sugar ele
vation in a diabetic.
Despite what the popular mediawould have us believe, fat is not
evil. In fact, many researchers are becomingquite concernedabout the
dangerous potential of "fat substitutes." Fat isabsolutely necessary for
survival. Much of the brain is constructed from fatty acids. Without
essential fatty acids — which, like essential amino acids, cannot be
manufactured by the body and must be eaten — you would die. Fat
substitutes such as the FDA-approved olestra (sold under the brand
name of Olean and present in such products as Frito-Lay WOW!
potato chips) bringabout the specter of people trying to subsist on a
no-fat diet, a diet that could kill them. (Olestraactuallyrobs the body
of important fat-soluble vitamins and essential fatty acids. The FDA
has required that it containadditives of those vitamins. In test mar
kets, some consumershavebeen made quite ill by the product, while
others don't see any effect. I don't recommend it — it's at best com
pletely unnecessary.)
Diabetics are affected disproportionately by diseases such as ather
osclerosis. This has led to the long-standingmyth that diabetics have
abnormal lipid profiles because they eat more fat than nondiabetics.*
It was likewiseonce thought that dietary fat caused all the long-term
complications of diabetes. For manyyears, this was takenasgospel by
most in the medicalcommunity. In truth, however, the high lipid pro
files in many diabetics with uncontrolled blood sugar havenothing to
do with the fatthey consume. Most diabetics consumevery little fat —
they've been conditionedto fear it. Highlipid profiles are a symptom
*Alipidprofile isthemeasuremenc of cholesterol, HDL (good cholesterol), LDL
(bad cholesterol), and triglyceride levels in the blood. Somephysicians nowcon
siderlipoprotein(a)to be an essential component of the lipidprofile. (See Chap
ter 2.)
TheBasic Food Groups 131
not of excess dietary fat, but of high blood sugars. Indeed, even in
most nondiabetics, the consumption of fat has little if anything to do
with their lipid profiles.
On the other hand, high consumption of carbohydrate, as we will
discuss shortly, can cause "nondiabetics" to develop someof the com
plications usuallyassociated with diabetes.
When I was on a verylowfat, high-carbohydrate diet about forty
years ago, I had highfasting triglycerides (usually over250mg/dl) and
high serum cholesterol (usually over 300 mg/dl), and I developed a
number of vascular complications. When I went onto a verylowcar
bohydrate diet and did not restrictmyfat,mylipidsplummeted.Now,
in mymidseventies, I have the lipidprofile of an Olympic athlete, ap
parently from eating a low-carbohydrate diet in order to normalize
mybloodsugars. That I exercise regularly probably doesn't hurt my
lipidprofile, either—but I was also exercising whenmylipidprofile
was abnormal.
Dare your physician. Askhim or her if his or her lipid profile on
a low-fat diet can remotely compare to mine, on a high-fat, low-
carbohydrate diet:
• LDL— the "bad" cholesterol — 53 (below 100 is considered
• HDL— the "good" cholesterol— 118 (above 39 is considered
• Triglycerides— 45 (below 150is considered normal)
• Lipoprotein(a) — undetectable (below 30 is considered normal)
Contraryto popular myth, fat is not a demon. It's the body'swayof
storing energy and maintaining essential organs such as the brain.
Without essential fattyacids, your bodywouldcease to function.
I've saved carbohydrate for last because it's the food group that ad
versely affects bloodsugarmost profoundly. Ifyou'relikemost diabet
ics—or virtually everyone who fives in an industrialized society —
you probably eat a diet that's mostly carbohydrate. Grains. Fruit.
Bread. Cake. Beans. Snack foods. Rice. Potatoes. Pasta. Breakfast ce
real. Bagels. Muffins. They look different, but dietarily speaking,
they're essentiallythe same.
132 Treatment
If you are already obese, you know and I know that you crave—
and consume — these foods and probablyavoid fats. As studies show,
you would be betteroff eating the fat than the carbohydrate. Fat alone
will be burned off. A combinationof high-carbohydrate foods and fat
will foster fat storage.
It is, therefore, amyth that Americans areoverweight due to exces
sive fat consumption. Americans are fat largely because of sugar,
starches, and other high-carbohydrate foods.
Accordingto statistics released by the U.S. Department of Agricul
ture, added sugar consumption hit an all-timehigh in 1999 (the last
yearforwhich statistics wereavailable), at awhopping 158 pounds per
American per year, an increase of 30 percentover 1983. The key word
here is"added." This doesn't account for starches and sugars naturally
present in food. According to a report from the Oregon Health Sci
encesUniversity, a 12-ounceStarbucksGrandeCaramel Mocha drink
contains 45 teaspoons of added sugar.
This increase in sugar consumption not coincidentallycorresponds
with the timing of recommendations to eat less fat. It was 1984 when
the National Institutes of Health (NIH) began advising everyone
within shouting distance to cut fat intake. It also corresponds quite
neatly with the creation of a whole new, multibillion-dollar industry
in low- and nonfat foods, many of which areextremely high in sugar.
Formore than ten years, the government had plannedto issue areport
once and for all damning fat asthe demon some scientistswere sure it
was. The problemwas, researchers couldn't"reverse engineer"the ac
tual data to make the science fit the assumption. Unfortunately, the
program to indict fat was left to dieaquiet death, andnot so much as
a press release wasissuedto say, "We were wrong." And so many of us
still don't know the truth. Theywerewrong.
No doubt the popular media have made you aware of the endless
procession ofbooks anddietsandadvertisements for foods alltouting
the valueof high"complexcarbohydrate" in the diet. Athletes"carbo-
load"beforebiggames or marathons. TV andradio commercials extol
the virtues of BrandX sports drink over BrandY because it contains
more "carbos."
As stunning as it sounds — and unbelievable, given the popular
media's recent love affairwith a high "complex carbohydrate,"low-fat
diet — you can quite easily survive on a diet in which you would eat
no carbohydrate. There are essential amino acids and essential fatty
acids,but there is no such thing asan essential carbohydrate. Further-
TheBasicFoodGroups 133
more, by sticking to a diet that contains no carbohydrate but has high
levels of fat and protein, you can reduce your cardiac risk profile —
serum cholesterol, triglycerides, lipoprotein(a), LDL, et cetera—
though you'd deprive yourself of all the supposed "fun foods" that we
cravemost.*We've allbeen trained to think that carbohydratesareour
best, most benign sourceof food, so how canthis be?
What if I, a physician, told you, adiabetic, to eat adiet that consisted
of 60 percent sugar, 20 percent protein, and 20 percent fat? More than
likely, you'd think I wasinsane.I'dthink I was insane,and I would never
makethis suggestion to adiabetic (norwouldI evenmake it to anondi
abetic). But this is just the diet the ADA recommended to diabetics for
decades. On the surface, these recommendations seemed to make sense
because of kidney disease, heart disease, and our abnormal Upid pro
files. But this is what is known assingle-avenue thinking. It seemedlog
ical to insistthat dietaryintakeof proteinand fat be reduced,becauseno
one had looked at elevated blood sugars and the high levels of insulin
necessary to bring them down asthe possible culprits.
So if you eat very little fat and protein, what's left to eat? Carbohy
As I discoveredin my yearsof experimentation on myself, and then
in my medical training and practice, the real dietary problem for dia
betics is not only fast-acting carbohydratebut also large amounts of
anycarbohydrate. In eithercase, the result ishighblood sugars requir
ing largeamounts of insulin to try to contain them.
So what arecarbohydrates?
The technical answer is that carbohydratesarechains of sugar mol
ecules. The carbohydrates we eat are mostly chains of glucose mole
cules. The shorter the chain, the sweeter the taste. Some chains are
longer and more complicated (hence, "simple" and "complex" carbo
hydrates), having many links and even branches. But simple or com
plex, carbohydrates arecomposed entirelyof sugar.
"Sugar?" you might ask, holding up a slice of coarse-ground,seven-
grainbread."This is sugar?"
In aword, yes, at least after you digest it.
With a number of important exceptions, carbohydrates, or foods
derived primarily from plant sources that are starches, grains, and
* You'd also be missing the vitamins and other nutrients contained in low-
carbohydrate vegetables, so a zero-carbohydrate diet is not in my ball game.
134 Treatment
fruits, have the same ultimate effect on blood glucoselevels that table
sugar does. (The ADA has recognized officially that, for example,
bread is as fast-acting a carbohydrate astable sugar. But instead of is
suingarecommendationagainst eating bread, its responsehasbeen to
saythat tablesugaris thereforeokay,and canbe"exchanged"for other
carbohydrates. To me, this is nonsense.)Whether you eat apieceofthe
nuttiest whole-grain bread, drink a Coke, or have mashed potatoes,
the effect on blood glucose levels is essentially the same— blood
sugar rises, fast, and in proportionto carbohydrate content.
As noted in the introduction to this chapter, the digestion process
breaks each of the major food groups down into its basic elements,
and these elements arethen utilized by the body as needed. The basic
element of most carbohydrate foods is glucose. We usually think of
simple carbohydrates as sugars and complex carbohydrates as fruits
and grains and vegetables. In reality, most fruit and grain products,
and some vegetables, are what I prefer to talk about as"fast-acting"
carbohydrates. Our saliva anddigestive tract containenzymes that can
rapidly chop the chains down into free glucose. We haven't the en
zymes to break down some carbohydrates, such as cellulose, or "indi
gestible fiber." Still, our saliva canbreakstarches into the shorterchains
on contact and then convert those into pure glucose.
Pasta, which is often made from durum wheat flour and water (but
can alsobe made from plainwhite flour and eggyolks, or other vari
ants), has been touted as a dream food — particularly for runners
carbo-loading before marathons — but it quickly becomes glucose,
and can raiseblood sugarvery rapidly for diabetics.
In the type 2 diabetic with impaired phase I insulin response, it
takes hours for the phaseII insulin to catchup with the postprandial
levels of glucose in the blood, anddayafter day, duringthat time, the
high blood sugars canwreakhavoc. In the diabetic who injects insu
lin, there is a tremendous amount of (rarely successful) guesswork
involved in finding the proper dosage and timing of insulin to cover
a carbohydrate-heavy meal, and the injected insulin not only doesn't
work fast enough, it is highlyunpredictable when taken in large doses
in attempts to cover large amounts of carbohydrate (see Chapter 7,
"The Laws of Small Numbers").
Some carbohydrate foods, like fruit, contain high levels of simple,
fast-acting carbohydrates. Maltose and fructose — malt sugar and
fruit sugar — for example, are slower-acting than sucrose — table or
cane sugar— but they will cause the sameincrease in blood sugarlev-
TheBask FoodGroups 135
els. It may be the difference between nearlyinstant elevation and ele
vation in 2 hours, but the elevation is still high, and a lot of insulin is
still required to bring it into line. And, if the insulin is injected, there's
the further problemof guesswork in timing and dosage. Despitethe
old admonition that an apple a day keeps the doctor away, I haven't
had fruit since 1970, and I am considerably healthier for it. Some
whole-plant vegetables, that is,those that come mostly fromthe stalks
and leaves, are of value to the diabetic and nondiabetic alike because
they contain considerable amounts of vitamins, minerals, and other
nutrients. (The recipe section of this book shows you a number of
tastyandsatisfying ways to work thesevegetables into your diet.)
As noted previously, most Americans who are obese are overweight
not because of dietary fat, but because of excessive dietary carbohy
drate. Much of this obesity is due to "pigging out" on carbohydrate-
richsnackfoods or junk foods, or evenon supposedhealthyfoodslike
whole-grain breadand pasta. It's my beliefthat this pigging out haslit
tle to do with hunger and nothing at all to do with beinga pig.
I'm convincedthat peoplewho crave carbohydrate have inherited
this problem. Tosome extent,we all have anatural craving for carbo
hydrate — it makes us feel good. The more people overeat carbohy
drates, the more they will become obese, evenif they exercise alot. But
certain people have a natural, ovenvhelming desire for carbohydrate
that doesn't correlate to hunger. These peoplein all likelihood have a
genetic predisposition toward carbohydrate craving, as well as a ge
netic predisposition towardinsulin resistance and diabetes. (See page
185, "The Thrifty Genotype.") This craving can be reduced for many
by eliminating such foods from the diet and embarkingupon a low-
carbohydrate diet.
In light of the above, you might guess that I advocate a no-
carbohydrate diet. In fact, in the next chapter you'll discover that I in
clude small amounts of carbohydrate in my meal plan. Backin 1970,
asI was still experimenting with blood sugarnormalization, I remem
beredthat during the twentieth century a newvitamin had been dis
covered every fifteen years or so. While theremay be no suchthing as
an essentialcarbohydrate, it seemed reasonable to conclude that, since
our prehistoric ancestors consumed some plants, plant foods might
well contain essential nutrients that were not yet present in vitamin
supplements and had not even been discovered. I therefore added
small amounts of low-carbohydrate vegetables (not starchy or sweet)
to my personal meal plan. All of a sudden I was eating salads and
136 Treatment
cooked vegetables instead of the bread, fruit, cereal, skim milk, and
pastathat I had been eatingon my prior ADA diet. It took a while to
get usedto salads, but nowI relish them. Only recently, in my lifetime,
have phytochemicals (essential nutrients found in plant foods) been
discovered. Phytochemicals are now incorporated into some vitamin
pills, but research on the use of isolated phytochemicals is still in its
early stages. You may haveheard of such phytochemical supplements
aslutein, lycopene, and soon. It would appear that many chemicals —
large numbers of which are likelynot even known about yet — work
together to provide beneficial effects. So at this point, it certainly
makes senseto eat low-carbohydrate salads and vegetables. (Although
fruits contain the same phytochemicalsasvegetables, they aretoo high
in fast-acting carbohydrate to be part of a restricted-carbohydrate
diet, asthe next chapterwill explain.)
Alcohol can provide calories, or energy, without directlyraisingblood
sugar, but if you'reaninsulin-dependent diabetic, you need to be cau
tious about drinking. Ethyl alcohol, which is the activeingredient in
hard liquor, beer, and wine, has no direct effect on blood sugar be
causethe body does not convert it into glucose. In the caseof distilled
spirits and very dry wine, the alcohol generally isn't accompanied by
enoughcarbohydrate to affect yourbloodsugar verymuch. For exam
ple, 100 proof gin has 83 calories per ounce. These extra calories can
increase your weight slightly, but not your blood sugar. Different
beers— ales, stouts, and lagers — can have varying amounts of car
bohydrate, which is slowenoughin its actionthat if you figure it into
your meal plan, it may not raise your blood sugar. Mixed drinks and
dessert wines can be loadedwith sugar, so they're best avoided. Excep
tions would be a dry martini or mixed drinks that canbe made with a
sugar-free mixer, suchassugar-free tonicwater.
Ethyl alcohol, however, can indirectly lower the blood sugars of
some diabetics if consumed at the time of a meal. It does this by par
tially paralyzing the liver and thereby inhibiting gluconeogenesis so
that it can't convert enough protein from the meal into glucose. For
the average adult, this appears to be a significant effect with doses
greater than 1.5 ouncesof distilled spirits, or one standard shot glass.
If you havetwo 1.5-ounce servings of ginwith ameal,your liver's abil-
The Basic Food Groups 137
ity to convert proteininto glucose maybe impaired. If you're insulin-
dependent andyourcalculation of howmuchinsulin you'll require to
cover your meal is based on, say, two hot dogs, and those hot dogs
don't get 7.5 percent converted to glucose, the insulin you've injected
will takeyour blood sugar too low. You'll have hypoglycemia, or low
blood sugar.
The problem of hypoglycemia itself is arelatively simple matter to
correct — you just eat some glucose and your blood sugar will rise.
But this gets youinto the kindof messy jerking up anddownof your
blood sugar that can cause problems. It's best if youcan avoid hypo-
andhyperglycemia (high bloodsugar) entirely.
Anotherproblemwith alcohol andhypoglycemia isthat if youcon
sume much alcohol, you'll have symptoms typical of both alcohol
intoxication and hypoglycemia — light-headedness, confusion, and
slurring of speech. Theonlyway you'll knowthecause of your symp
toms isif you've been monitoring your blood sugar throughout your
meal. This isunlikely. So youcould find yourself thinking you've con
sumed too much alcohol when in fact your problem is dangerously
lowblood sugar. In such asituation, it wouldn't even occur to youto
check your blood sugar. Remember, that earlyblood sugar-measuring
device I gotwas developed in order to help emergency roomstaffs tell
the difference between unconscious alcoholics and unconscious dia
betics. Don't make yourself an unconscious diabetic. A simple over
sight could turn fatal.
Many of the symptoms of alcohol intoxication mimic those of ke
toacidosis, or the extreme highbloodsugar andketone buildupin the
bodythat can result in diabetic coma. The great buildup of ketones
causes a diabetic's breath to have an aroma rather like that of someone
who's been drinking. If youdon't die of severe hypoglycemia, thenyou
mighteasily dieof embarrassment when youcome to andyourfriends
are aghast and terrified that the emergency squad had to be called to
bring you around.
Insmall amounts, alcohol isrelatively harmless — oneglass of dry
wine or beer with dinner — but if you're the type who can't limit
drinking, it's best toavoid it entirely. For thereasons already discussed,
and contrary to the guidelines of the ADA, alcohol can be more be
nign between meals than it is at meals. One benevolent effect of alco
hol is that it can enable some diabetics to consume one beer or one
small bloody Mary (tomato juice mixed with an ounce and a half of
vodka) without raising blood sugar.
Diet Guidelines Essential to the
Treatment of All Diabetics
Researchinto creatingreplacement ceUs for burned-out insulin-
producing pancreatic beta ceUs is so promising that it's
temptingto think of a"cure" not in terms of if but when. The
reaUty is, however, less rosy. There may one daybe a cure, but to put
off normalizing your blood sugars until then is simply to ignore the
reaUty of your situation. If you're going to controlyour diabetes and
get on witha normal life, youwiU have to change your diet, and the
when is now. No matter how mild or severeyour diabetes, the key as
pect of aU our treatment plans for normalizing bloodsugars and pre
ventingor reversing complications of diabetes is diet. In the terms of
the Laws of SmaU Numbers, the singlelargest"input" you can control
is what you eat.
The next several pages may weU be the most difficult pages of this
book for you to accept —as weU as some of the most important.
They're fuU of the foods you're going to have to restrict or eliminate
from your diet if you're going to normalize your blood sugars. You
may see some ofyourfavorite foods onour No-No Ust, but before you
stop reading, keep in mind a few important things. First, towardthe
end of this chapter we discuss the foods you can safely eat. Second,
whUe you wiU have to eliminate certain foods, there are some gen
uinely sugar-free andlow-carbohydrate alternatives.
One purpose of blood glucose self-monitoring is to learn through
Diet Guidelines Essential to theTreatment ofAllDiabetics 139
your blood sugar profiles how particular foods affectyou. Bloodsugar
self-monitoring is the ultimate measure of the effect foods have on
your blood sugar. If you don't beUeve what you'rereading here, check
your blood sugars every 2 hours afterconsuming food you arecertain
must be benign. Over yearsof examining profileslike the ones you wiU
create, I've observed that some people are more tolerant of certain
foods than other people. For example, bread makes my own blood
sugar risevery rapidly. Yet one or two of my patientswith mUdtype 2
diabetes eat asandwichof thin breadeverydaywith only minor prob
lems. InevitablyI findthis is related to delayed stomach-emptying (see
Chapter 22). In any case, you should feel free to experiment with food
and then perform blood sugar readings. It's likely that for many dia
betics most or aU of our restrictions wiU be necessary.
Patients often ask, "Can't I just take my medication and eat what
ever I want?"It almost seems logical, and would be fine if it worked.
But just takingyour medicationandeating whatever you want doesn't
work — because of the Laws of SmaU Numbers — so we have to find
something that does.
Manydiabetics canbe treatedwith diet alone,and if your disease is
relatively mUd, you could easUy faU into this category. Some patients
who havebeen using insulin or oral agents find that once on our diet
theyno longer needbloodsugar-lowering medication. But evenif you
require insulin or other agents, diet wiU still constitute the most es
sential part of your treatment.
Think smaU inputs. You may recaU from prior chapters that — for
even the nuldest diabetic— the impairment or loss of phase I insulin
responsemakes normalizing blood sugars impossible for at least a few
hours after a high-carbohydrate meal. Eating even smaU amounts of
fast-acting carbohydrate raises blood sugar sorapidly that anyremain
ingphase II insulinresponse cannot promptlycompensate. This istrue
if you're injectinginsulinor if you're still makingyour own insulin.
Any sensiblemeal plan for normalizing blood sugartakes this into
account and foUows these basic rules:
• First, eliminate aU foods that contain simple sugars. As you
should knowby now — but it bearsrepeating— "simple sugar"
does not mean just table sugar; that's why I prefer to caU them
fast-acting carbohydrates. Breads and other starchy foods, such
as potatoes and grains, become glucose so rapidly that they can
cause serious postprandial increases in blood sugar.
140 Treatment
• Second, limit your total carbohydrate intake to an amount that
wiU work with your injected insuUn or your body's remaining
phase II insulin response, if any. In this way, you avoid a post
prandial blood sugar increase, and avoid overworking any re
maining insulin-producingbeta ceUs of your pancreas (research
has demonstrated that beta ceU burnout can be slowed or halted
by normalizingblood sugars).
• Third, stop eating when you no longer feel hungry, not when
you're stuffed. There'sno reasonfor you to leave the table hun
gry, but there's also no reason to be gluttonous. Remember the
Chineserestaurant effect (page97).
• FinaUy, for best results, foUow a predetermined meal plan (see
next chapter).
Sometimes you'U find yourself at a restaurant, hotel, or reception
where you cannot predict if foods havesugar or flour in them. Your
waiter probably has Uttle idea of what's in a givenrecipe, so don't even
ask him; his response wiU likely be incorrect. I've found that the easi
est way to make certain is to use the Clinistix or Diastix that should
havebeen checkedoff on your supplyUst (Chapter 3). These are man
ufactured to test urine for glucose. Weuse them to test food. If, for ex
ample, you want to determine if a soup or salad dressing contains
tablesugar (sucrose) or a saucecontainsflour, just put a small amount
in your mouth and mixit withyour saUva. Thenspit a tinybit onto a
test strip. Any color change indicates the presence of sugaror starch.
SaUva is essential to this reactionbecause it contains an enzyme that
releases glucose from sucrose (table sugar) or from flour in the food,
permitting it to react with the chemicals in the test strip. This is how I
found that one restaurant in myneighborhooduseslargeamounts of
sugar in its bouiUonwhUe another restaurant uses none.*
Solid foods can also be tested this way, but you must chewthem
*I usethis test on television to showthat even"wholegrain"breads,contrary to
claimsof the ADA, becomeinstant glucose when exposedto saliva.
Diet Guidelines Essential to theTreatment ofAllDiabetics 141
first The Ughtest color on the color chart label of the test strip con
tainer indicates a very low concentration of glucose. Any color paler
than this may be acceptable for foods consumed in small amounts.
The Clinistix/Diastix method works on nearly aU the foods on our
No-NoUst exceptmilk products, whichcontainlactose. It wiU alsonot
react with fructose (fruit sugar; alsopresent in some vegetables and in
honey). If in doubt, assume the worst.
Named beloware some of the common foods that contain simple sug
ars, which rapidly raise blood sugar or otherwise hinder blood sugar
control and should be eliminatedfromyour diet. AU grain products,
for example — fromthe flourin"sugar-free" cookies to pastato wheat
or non-wheat grain products except pure bran — are converted so
rapidlyinto glucoseby the enzymes in saUva and further down in the
digestive tract that they are, as far as blood sugar is concerned, essen
tially no different than table sugar or even pure glucose. There are
plenty of food products, however, that contain such tiny amounts of
simple sugars that they wiU have a negUgible effect on your blood
sugar.One gram of carbohydratewillnot raiseblood sugar more than
5 mg/dl for most diabetic adults (but considerably more for small
children). A single stick of chewing gum or a single tablespoon of
salad dressing made with only 1 gram of sugar certainly poses no
problems. In these areas, you have to use your judgment and your
blood sugar profiles. If you're the type who, once you start chewing
gum, has to have a newstick every30minutes, then you should prob
ably avoid chewing gum. If you have delayed stomach-emptying
(Chapter 22), smaU amounts of "sugar-free" chewinggum may help
faciUtate your digestion.
Powdered Artificial Sweeteners
At this writing, several artificial sweeteners are avaUable. They are
avaUable from different manufacturers under different names, and
some, such as Equal and Sweet'n Low, can have brand names under
which more than one form of sweetener is sold. Here, to simpUfyyour
shopping, are acceptable products currently and soon to be avaUable:
142 Treatment
saccharintablets or Uquid(Sweet'n Low)
aspartame tablets (Equal, NutraSweet)*
acesulfame-K (Sunett, The Sweet One)
steviapowderor Uquid (stevia hasnot been approved in the
sucralose tablets (Splenda)
neotame tablets — when avaUable (newlyapproved by the FDA)
cyclamate tablets andUquid (not yet avaUable in the UnitedStates)
These are aU noncarbohydrate sweeteners that vary in their avaU-
abiUty andcanbe usedto satisfyasweet tooth without significantly af
fecting blood sugars. But when sold in powdered form, under such
brand names asSweet'n Low, Equal, The Sweet One, Sunett, Sugar Twin,
Splenda, andothers, these products usually contain a sugar to increase
bulk, andwill rapidly raise blood sugar. They are aU orders of magni
tude sweeter tastingthan sugar. When you buy them in packets and
powdered form, with the exception of stevia, they usuaUy contain
about 96 percent glucose or maltodextrin and about 4 percent artifi
cial sweetener. If you read the "Nutrition Facts" label on granulated
Splenda, for example,it Usts, as suchlabels must, ingredientsin order
from most to least: dextrose (glucose), maltodextrin (a mixture of
sugars), and finally sucralose. Most powdered sweeteners are sold as
low-calorie and/or sugar-free sweeteners because they containonly 1
gramof a sugar ascompared to 3 grams of sucrose in a similar paper
packet labeled"sugar." More suitable for diabetics aretablet sweeten
ers such as saccharin, cyclamate, and aspartame. As noted above, the
same brand name can denote multiple products: Equal is a powder
containing 96percent glucose andalso atablet containing aminuscule
(acceptable) amount of lactose. Sweet'n Lowpowderis saccharin with
96 percent glucose. Stevia powder and liquid (sold in health food
stores) contain no sugar of anykind and only minute amounts of car
Anew"natural artificial" sweetener, caUed tagatose (nobrandname
asof this writing), hasbeen approvedfor sale in the United States. De
rived frommilk, it'sclaimed to be 92percent assweetassugar, with no
* ManyWebsites falsely perpetuate the myth that aspartame is toxicbecause its
metaboUsm produces the poison methanol. In reality, one 12-ouncecan of an
aspartame-sweetened soft drink generates only Vis as much methanol as does a
glassof milk.
Diet Guidelines Essential to the Treatment ofAll Diabetics 143
aftertaste and no effect on blood sugars. This last claim — that it has
no effect on blood sugars— remainsto be seen.In many cases, what's
termed "no effect" or "negUgible effect" usuaUy has significantenough
effectto make blood sugar control difficult.
Another newartificial sweetener, neotame, is being sold as an addi
tivebythe makers of NutraSweet. It is supposedly8,000times as sweet
as table sugar. Its use as a food additiveshould pose no problems, but
if it becomes avaUable to consumers as a powder, it wiU probably be
mixed with a sugar as in the instances cited above.
Yet another new powdered sweetener, erythritol (Zsweet) is pro
moted as being 70 percent as sweet as table sugar, but to my taste it is
much less sweet, so that a considerable amount must be used. Since
erythritol is a sugar alcohol, it will raise diabeticblood sugars signifi
cantlywhen consumed by the tablespoon, as I found necessary.
So-Called Diet Foods and Sugar-Free Foods
BecauseU.S. food-labeling laws in the recent past have permitted and
thus encouraged products to be caUed "sugar-free"if they do not con
tain common table sugar (sucrose), the mere substitution of another
sugar for sucrose has permitted the packagerto deceive the consumer
legaUy. Most so-caUed sugar-free products havebeen, for many years,
fuU of sugars that may not promote tooth decaybut most certainly
wiU raise your blood sugar. If you'vebeen deceived, you're not alone.
I've been in doctors' officesthat have candy dishes fuU of"sugar-free"
hard candies for their diabetic patients! Sometimes the label wiU dis
close the name of the substitute sugar.
Here is a partial list of some of the many sugars you can find in
"sugar-free" foods. AU of thesewiU raiseyour blood sugar.
carob honey saccharose
corn syrup lactose sorbitol
dextrin levulose sorghum
dextrose maltodextrin treacle
dulcitol maltose turbinado
fructose mannitol xyUtol
glucose mannose
Some, such as sorbitol and fructose, raise blood sugar more slowly
than glucosebut stiU too much and too rapidly to prevent a postpran
dial blood sugar rise in people with diabetes.
144 Treatment
Other "diet" foods contain either sugars that are alternates to su
crose, large amounts of rapid-acting carbohydrate, or both. Many of
these foods (e.g., sugar-free cookies) are virtuaUy 100percent rapid-
actingcarbohydrate, usuaUy flour, so that evenif theywereto contain
none of the above added sugars, consumption of a smaU quantity
would easUy causerapid blood sugar elevation.
There are exceptions:
• Most diet sodas — with some glaringexceptions,so always check
nutrition labels and look for 0 under carbohydrate.* So-caUed
sugar-freeSlice contains 40percent "natural fruit juice"
• Sugar-free JeU-0 brand gelatindesserts — the ready-to-eat vari
ety, not the powdered mix(see page 161)f
• DaVinci brand sugar-free syrups (seepage 159)
AU of these are made without sugar of anykind. These you need not
restrict. See"SoWhat's Left to Eat?"later in this chapter.
Candies, Including "Sugar-Free" Brands
Atiny "sugar-free" hard candy containing only 2Vi grams of sorbitol
can raise blood sugar almost 13 mg/dl. Ten of these can raise blood
sugar 125 mg/dl. Since sorbitol, for example, has only one-third the
sweetening power of sucrose, the manufacturer uses three times as
much to get the same effect. This wiU raise blood sugars three times
as much as, although more slowlythan, table sugar.
Honey and Fructose
In recentyearsa number of"authorities" have claimedthat honey and
fructose (a sugar occurring in fruits, some vegetables, and honey) are
useful to diabetics because they are "natural sugars." WeU, glucose is
themost natural of the sugars, since it is present in aU plants and aU
but one known species of animal, and we alreadyknow what glucose
* Looking for 0 under carbohydrate may not tell you everything you want to
know.Alsolook in the Ust of ingredientsto seeif the product contains any of the
sugars Usted. If it does, checkyour blood sugars after drinking, if you choose to
drink them, and seewhat effectthey haveon you.
f Unfortunately, the manufacturers of sugar-free Jell-O brand gelatin recendy
started to add maltodextrin to the powderedversion. I expect that they will soon
add it also to the ready-to-eat version.Asuitable substitute would be Knox un-
flavored gelatinwith added liquidsteviaand your choiceof Da Vinci sugar-free
syrup or WaterSensations for flavoring.
Diet Guidelines Essentialto the Treatmentof All Diabetics 145
can do to blood sugars. Fructose, which is sold as a powdered sweet
ener, is often derivedfrom corn (a grain) and is a significant ingredi
ent in many food products (as in high-fructose corn syrup). Honey
and fructose, "natural" or not, wiU raise blood sugar far more rapidly
than either phase II insulin release, injected insulin, or oral hypo
glycemic agents can bring it down. Just eat a fewgrams of honey or
fructose and checkyour blood sugar every 15minutes. You wiU read-
Uy prove that "authorities" can be wrong.
Desserts and Pastries
With the possible exception of products marked"carbohydrate— 0"
on the nutrition label, virtuaUy every food commonly used for
desserts wiU raiseblood sugar too much and too fast. This is not only
because of added sugar but alsobecause flour, milk, and other com
ponents of desserts are veryhigh in rapid-actingcarbohydrate.
Bread and Crackers
One average sUce of white, rye, or whole grain bread contains 12 or
more grams carbohydrate.The"thin" or "lite"breads are usuaUy cut at
half the thickness of standard bread sUces and therefore contain half
the carbohydrate. So-caUed high-proteinbreads contain only a smaU
percentage of their calories as protein and are not significantly re
duced in carbohydrateunless they are thinly cut. Brownbread, raisin
bread, and corn bread aU contain as much or more fast-acting carbo
hydrate than rye, white, or whole wheat. Some diabetics with severe
gastroparesis (Chapter 22) can tolerate the inclusion of 1-2 sUces of
thin bread or a few crackers as part of their low-carbohydrate meal
limits. Unfortunately, most of us experience very rapid increases of
blood sugar after eating even smaU amounts of such products (bread,
crackers, cereals, pastryshells, et cetera) madefromanygrain. This in
cludes those made from less common grains, such as barley, kasha,
oats, sorghum, and quinoa.
Rice and Pasta
Both pasta and wUd rice (which is actuaUy not a true variety of rice
but another grain entirely) are claimedbysome nutrition authorities
to raiseblood sugar quite slowly. Just checkyour blood sugar levelsaf
ter eating them and you'Uagain prove the "authorities" wrong. Alter
natively, you might try the CUnistix/Diastix test described on pages
140-141. Like wUd rice and pasta, white and brown rices also raise
146 Treatment
blood sugar quite rapidly for most of us and should be avoided. The
same is true of rice cakes.
Breakfast Cereals
Most cold cereals, likesnack foods, are virtuaUy 100percent carbohy
drate, even those claiming to be "high protein." AdditionaUy, many
contain large amounts of added sugars. Since they are made from
grain, smaU amounts, even of whole-grain cereals, wiU cause a rapid
risein blood sugar (according to the glycemic index,a measureof how
rapidlyfoods are metabolized into glucose, brown rice actuaUy raises
blood sugar faster than white rice). Even bran flakes are mostlyflour.
If youhavebeeneatingbran flakes to improve bowel function,youcan
substitute verysmaU amounts (1 tablespoon) of psyUium husks pow
der, which is entirelyindigestible fiber. Useonly the sugar-freevariety
of MetamucU or other such products. (You can get the husks powder
at a health food store and mixwith water. If you don't care for the tex
ture or taste, you can drink it mixedin diet soda.) You can also make
your own cerealfrompure bran.
Cookedcereals generaUy containabout 10-25 grams of fast-acting
carbohydrate per half-cup serving. I find that even smaU servings
make blood sugar control impossible.
Snack Foods
These are the products in ceUophane bags that you find in vending
machines and supermarkets. Theyinclude not just candy,cookies, and
cakes, but pretzels, potato chips, taco chips, tiny crackers, and pop
corn. These foods are virtuaUy 100 percent carbohydrate and fre
quentlyhaveaddedsucrose, glucose (thelabel maysaydextrose), corn
syrup, et cetera. Althoughsome nuts (e.g., macadamia) are relatively
low in carbohydrate, who can sit down and eat only six macadamia
nuts (about 1gramof carbohydrate)? It's simpler just to avoidthem.
So-Called Protein Bars
Although drugstore and groceryshelves are fuU of bars that claim to
be "protein bars,"most are reaUy nothing more than candy bars with
"healthy" packaging. The FDA recently analyzed twenty different
brands and found that aU but two contained much more carbohydrate
than stated on the labels. These were removed from the marketplace,
but many more remain. This is another case of when it sounds too
good to be true, it probablyis.
Diet GuidelinesEssential to the Treatment ofAll Diabetics 147
Milk and Cottage Cheese
Milkcontains a considerable amount of the simplesugar lactoseand
wiU rapidly raise blood sugar. SkimmUk actuaUy contains more lac
toseper ounce than does wholemilk. One or 2 teaspoonsof milk in a
cup of coffee wiU not significantly affect blood sugar, but lA cup of
milk wiUmake a considerable difference to most of us. Cream, which
you have probablybeen instructed to avoid, is okay. One tablespoon
has only 0.5 gram of carbohydrate. Furthermore, it tastes much better
than substitutes and has considerably more "lightening power." The
powdered lighteners for coffee contain relatively rapid acting sugars
and shouldbe avoided if youusemorethan a teaspoonful at a time or
drink more than 1 cup of coffee at a meal. A coffee lightener worth
consideringis WestSoy brand soymilk, which is sold in health food
storesthroughout the UnitedStates. Although several WestSoy flavors
are marketed, only the ones marked 100% Organic Unsweetened are
unsweetened. It comes in plain and vanilla and usuaUy contains
5grams of carbohydratein 8ounces.Other unsweetened brands, such
asVitasoyandYu Natural, areavaUable invarious parts of the country.
One catch — soymilkcurdles in veryhot coffee or tea.
Cottage cheese also contains a considerable amount of lactose be
cause, unlike most other cheeses (hard cheese, cream cheese), which
are okay, it is onlypartlyfermented. I wasunaware of this untU several
patients showed me records of substantial blood sugar increases after
consuming a container of cottagecheese. It should be avoidedexcept
in very smaU amounts, sayabout 2 tablespoons.
Fruits and Fruit Juices
These contain varying mixtures of simple sugars and more complex
carbohydrates, aU of whichwiU act dramatically on blood sugar levels,
which youcan prove witha few experiments with blood sugar mea
surements. Bitter-tastingfruits such as grapefruit and lemon contain
considerable amounts of simple sugars. They taste bitter because of
the presence of bitter chemicals, not because sugar is absent. Orange
juice, which may be high in vitamin C, alsocontains about as much
sugar as a nondiet soft drink. Although eliminating fruit and fruit
juices fromthe diet can initiaUy be a bigsacrifice for manyof my pa
tients, they usuaUy get used to this rapidly, and they appreciate the ef
fect upon blood sugar control. I haven't eaten fruit in almost forty
years, and I haven't suffered in anyrespect. Some peoplefear that they
148 Treatment
wiU lose important nutrients by eliminating fruit, but that shouldn't
be a worry. Nutrients found in fruits are alsopresent in the vegetables
you can safely eat.
In our society, wegeneraUy reserve the name"fruit"for sweetfruits,
such as apples, oranges, and bananas, aU of which you should avoid.
There are, however, a number of biological fruits (the part of certain
plants that contains pulp and seeds) that are benign for the diabetic,
such as summer squash, cucumbers (includingmany types of pickle),
eggplant, beUand chUi peppers, and avocado. These tend to have large
amounts of ceUulose, an undigestible fiber, rather than fast-acting
carbohydrate. (It's worth noting that ceUulose, found in vegetables
and fruits, is essentially the same fiber that makes up much of the
shady elm on the corner. It has undigestible caloriesyour body won't
metaboUze becausewedon't havethe enzymesto break down the spe
cial ceUuose chains of sugarsinto digestible form.)
Beets. Likemost other sweet-tastingvegetables, beets are loaded with
sugar. Sugarbeets are a sourceof tablesugar.
Carrots. Aftercooking,carrots tastesweeter and appear to raiseblood
sugar much more rapidlythan when raw. This probablyrelatesto the
breakdown of complex carbohydrates into simpler sugars by heat.
Evenrawcarrots should be avoided. If, however, you are served a salad
with a fewcarrot shavings on top for decoration, don't bother to re
move them. The amount is insignificant, just like a teaspoon of milk.
Corn. Not a vegetable at aU but a grain, as noted above. NearlyaU of
the corn grown in the United States is used for two main purposes.
One is the production of sweeteners. Most of the sugar in Pepsi-Cola,
for example, comes from corn. The other major purpose is animal
feed, e.g., fatteningup hogs, cattle, and chickens. Corn for consump
tion by people, as a"vegetable" or as snackfoods, comesin third. Dia
betics should avoid eating corn, whether popped, cooked, or in
chips — even 1 gram of corn (a couple of kernels of popcorn) wiU
rapidly raise myblood sugar byabout 5 mg/dl.
Potatoes. For most diabetics, cooked potatoes raise blood sugar al
most as fast as pure glucose, even though they may not taste sweet.
Diet Guidelines Essential to the Treatment ofAll Diabetics 149
Giving up potatoes is a big sacrificefor many people, but it wiU also
make a big differencein your postprandial blood sugars.
Tomatoes, tomato paste, and tomato sauce. Tomatoes, as you
know, are actuaUy a fruit, not a vegetable, and as with citrus fruits,
their tangcanconceal just howsweet theyare. Theprolonged cooking
necessary for the preparation of tomato sauces releases a lot of glu
cose, and you would do weU to avoid them. If you're at someone's
home for dinner and are served meat or fish covered with tomato
sauce, just scrape it off. The smaU amount that might remain should
not significantly affect your bloodsugar. If you are having them un
cooked in salad, limit yourself to one sUce or a single cherry tomato
per cup of salad. (See page 394 for a recipe for a low-carbohydrate,
tomato-free, ItaUan-style red sauce that can be good over, say, a
broUed, sautied, or griUed chicken breast or veal scaUopine.) Onions
faU into this same category— despitesomesharp flavor, they're quite
sweet, some varieties sweeter than others. There are other vegetables
in the alUum famUy that canbe easUy substituted, althoughin smaller
quantities,such as shallotsand elephant garlic.
Commercially prepared soups. BeUeve it or not, most commercial
soups marketedin this country canbe asloadedwith added sugar as a
soft drink. The taste of the sugar is frequently masked by other fla
vors — spices,herbs, and particularly salt. Evenif there were no added
sugar, the prolonged cooking of vegetables can breakthe special glu
cose bonds in the ceUulose of slow-acting carbohydrates, turning
theminto glucose. As youknowfrom above, the amount of carbohy
drateclaimed on the Nutrition Facts label canvaryconsiderably from
what's actuaUy in the can.Addto that the commoninclusion of pota
toes, barley, corn, rice, and other unacceptable foods, and you havea
product that you shouldavoid. There arestill somecommercial soup
possibilities that fit into our scheme. See the corresponding heading
on page 154.
Health foods. Of thehundreds of packaged food products that you
seeon the shelves of the average healthfoodstore, perhaps 1percent
are lowin carbohydrate. Many are sweetened, usuaUy with honeyor
other so-caUed natural sugars. Indeed, many so-caUed natural foods
canbeveryhighin carbohydrate. Since thehealthfoodindustryshuns
artificial (nonsugar) sweeteners like saccharin or aspartame, if a food
150 Treatment
tastes sweet, it probablycontains asugar. There area few foods carried
by these stores that are unsweetened and lowin carbohydrate. You'U
find some of these Usted laterin this chapter.
It's agoodquestion, andthe same one I asked myselfnearly fortyyears
ago as I discovered that more andmore of the things that the Ameri
canDiabetes Association hadbeentellingme wereperfectlyfineto eat
made blood sugar control impossible. In the foUowing pages, I'll give
you abroad overview of the kinds of food my patients andI usuaUy
eat. Please remember that with the exception of the no-calorie bev
erages (including seltzer water and mineral water with no added
carbohydrate) and moderate portions of sugar-free JeU-O without
maltodextrin, there areno "freebies." VirtuaUy everything we eat wiU
have some effect upon blood sugar if enough is consumed. You may
discover things I've neverheard of that havealmost no effect on your
blood sugar. If so, feel free to include them in your meal plan, but
checkyourbloodsugar every halfhour for a few hours before assum
ing that they arebenign.
Most vegetables, other thanthoseUsted in the No-No section, are ac
ceptable. Acceptable vegetables include asparagus, avocado, broccoli,
brussels sprouts,cabbage andsauerkraut, cauliflower, eggplant, onions
(in smaU amounts), peppers (any color except yeUow), mushrooms,
spinach, stringbeans, summersquash, andzucchini. Asarule ofthumb,
% cup of whole cooked vegetables, Vi cup of diced or sUced cooked
vegetable, lA cup mashed cooked vegetable, or 1 cup of mixed salad
acts upon blood sugar asif it contains about 6 grams of carbohydrate.
Remember that cooked vegetables tend to raise blood sugar more
rapidly than raw vegetables because the heat makes them more di
gestible andconverts some of the ceUulose to sugar. GeneraUy, more
cookedvegetables by weight wiU occupyless volume in a measuring
cup, so a cup of cookedspinach wiU weigh considerably more than a
cup of uncooked. On yourself-measurements, note howyour favorite
vegetables affect your blood sugar. Raw or unmashed vegetables can
present digestive problems to people with gastroparesis.
Of the foUowing cookedvegetables, eachactsupon blood sugar as
Diet Guidelines Essential to theTreatment ofAllDiabetics 151
if it containsabout 6 grams of carbohydrate in %cup (aU cooked ex
cept as noted):
artichoke hearts
bamboo shoots
beet greens
beU peppers (green
and red only, no
yeUow) (cooked
or raw)
bok choy (Chinese
celeryroot (celeriac)
coUard greens
daikon radish
dandeUon greens
hearts of palm
mustard greens
patty pan squash
snow peas
spaghetti squash
summer squash
turnip greens
water chestnuts
zucchini flowers
In addition to the above, youshould keepthe foUowing in mind:
• Onions are high in carbohydrate and should only be used in
smaU amounts for flavoring — smaU amounts of chives or shal
lots can pack alot of flavor.
• One-halfsmaU avocado contains about 6grams of carbohydrate.
• One cup mixed green salad without carrots and with a single
slice of tomato or onion has about the same impact on blood
sugarsas6 grams of carbohydrate.
• One-quarter cup mashed pumpkin contains about 6 grams of
carbohydrate. My own opinion is that without some flavoring,
pumpkin tastes about as appetizing as Kleenex. Therefore I flavor
it with much steviaandspice(cinnamon) and warmit to make it
abit likepumpkin pie filling. (For othervegetables from this list,
suchas turnips, assume that lA cup of the mashed product con
tains 6 gramsof carbohydrate.)
Meat, Fish, Fowl, Seafood, and Eggs
These are usuaUy themajor sources of calories inthemeal plans of my
patients. The popular press is currently down on meat andeggs, but
my personal observations andrecent research impUcate carbohydrates
152 Treatment
No-No's in a Nutshell
Here is a concise Ust of foods to avoid that are discussed in this
chapter.You maywant to memorizeit or copyit, as it is worth
Sweets and Sweeteners
• Powdered sweeteners (other than stevia)
• Candies, especiaUy so-caUed sugar-freetypes
• Honey and fructose
• Most "diet"and "sugar-free" foods (except sugar-free
JeU-0 gelatin whenthe label doesn't mention malto
dextrin, and diet sodas that do not contain fruit juices
or Ust other carbohydrate on the label)
• Desserts (exceptJeU-0gelatinwithout
maltodextrin — no more than xh cup per serving) and
pastries: cakes, cookies, pies,tarts, et cetera
• Foods containing, as a significantingredient, products
whose names end in -olor -ose (dextrose,glucose, lac
tose, mannitol, mannose, sorbitol, sucrose, xylitol, xy
lose,et cetera),except ceUulose; also, corn syrup,
molasses, maltodextrin, et cetera
Sweet or Starchy Vegetables
• Beans: chili beans, chickpeas,lima beans, lentils, sweet
peas, et cetera (stringbeans, snowpeas, and beU and
chiU peppers,whichare mostlyceUulose, are okay, as
are verylimitedamounts of manysoybeanproducts)
• Beets
• Carrots
• Corn
• Onions, exceptin smaU amounts
• Packaged creamedspinachcontainingflour
• Parsnips
• Potatoes
Diet Guidelines Essentialto the Treatmentof All Diabetics 153
• Cooked tomatoes, tomato paste, tomato sauce, and
raw tomatoes except in smaU amounts
• Winter squash
Fruit and Juices
• AU fruits (except avocados)
• AU juices (includingtomato and vegetable juices—
exceptfor some people,in a smaU BloodyMary)
Certain Dairy Products
• Milk
• Sweetened, flavored, and low-fat yogurts
• Cottage cheese (except in very smaU amounts)
• Powderedmilksubstitutesand coffee Ughteners
• Canned milk concentrate
Grains and Grain Products
• Wheat, rye, barley,corn, and lesser-known,"alterna
tive"grains, such as kasha, quinoa, and sorghum
• White, brown, wUd rice, or rice cakes
• Pasta
• Breakfast cereal
• Pancakes and waffles
• Bread, crackers, and other flour products, including
"whole grain" breads
Prepared Foods
• Most commerciaUy prepared soups
• Most packaged"health foods"
• Snackfoods (virtuaUy anything that comes wrapped
in ceUophane, includingnuts)
• Balsamic vinegar (comparedto winevinegar, white
vinegar, or cider vinegar, balsamic contains consider
able sugar)
154 Treatment
rather than dietaryfat in the heart disease and abnormal blood Upid
profiles of diabetics and even of nondiabetics. If youare frightenedof
these foods, you can restrict them, but deprivingyourself wiU be un
likely to buyyouanything. Appendix AdetaUs the current controversy
and the shakyscience behind the present, faddish high-carbohydrate
dietary recommendations, and lays out my concerns and opinions.
Eggyolks, by the way, are a major source of the nutrient lutein, which
is beneficial to the retina of the eye. Organic eggs contain large
amounts of omega-3 fattyacids, whichare goodfor your arteries.
Tofu, and Soybean Substitutes for Bacon, Sausage,
Hamburger, Fish, Chicken, and Steak
About half the calories in theseproducts come from benevolent veg
etable fats, and the balance from varying amounts of protein and
slow-acting carbohydrate. They are easy to cook in a skiUet or mi
crowave. Protein and carbohydrate content should be read from the
labels and counted in your meal plan. Their principal value is for
people who are vegetarian or want to avoid red meat. Health food
stores stock many of these products. For the purpose of our meal
plans, as described in the next chapter, remember to divide the grams
of protein listed on the packageby 6 in order to get"ounces" of protein
(seepage 171).
Certain Commercially Prepared and Homemade Soups
Although most commercial and homemade soups contain large
amounts of simple sugars, you can learn how to buy or prepare low-
or zero-carbohydratesoups (seesuggestions below).Manybut not aU
packaged bouillon preparations have no added sugar and only small
amounts of carbohydrate. Checkthe labelsor use the Clinistix/Diastix
test, observing the special technique described on pages 140-141.
Plain consomme' or broth in some restaurants may occasionaUy be
prepared without sugar. Again, checkwith Clinistix/Diastix.
Homemade soups, cooked without vegetables, can be made very
tasty if they are concentrated.You can achieve this by barelycovering
the meat or chickenwith water whUe cooking, rather than fiUing the
entire pot with water, as is the customaryprocedure. Alternatively, let
the stock cook down (reduce) so you get a more concentrated, flavor
ful soup. You can alsouse herbs and spices, aU of which have negligi
ble amounts of carbohydrates, to enhance flavor. See "Mustard,
Pepper, Salt, Spices, Herbs," page 160. Clam broth (not chowder) is
Diet Guidelines Essential to theTreatment ofAll Diabetics 155
usuaUyverylowin carbohydrate.In the UnitedStatesyou can alsobuy
clamjuices (not Clamato), which contain only about 2 grams of car
bohydrate in 3 fluid ounces. CampbeU's canned beef bouiUon and
consomme' contain only 1gramcarbohydrateper serving. CoUege Inn
brand canned chickenbroth contains no carbohydrate.Most bouiUon
cubesare alsolowin carbohydrate; readthe labels.
Cheese, Butter, Margarine, and Cream
Most cheeses (other than cottagecheese) containapproximately equal
amounts of protein and fat and smaU amounts of carbohydrate. The
carbohydrateand the protein must be figured into the meal plan, as I
wiU explain in Chapter 11. For people who want (unwisely) to avoid
animal fats, there are some special soybean cheeses (not very tasty).
There's also hemp cheese, which I knownothing about. Cheese is an
exceUent source of calcium. Every ounce of whole milk cheese con
tains approximately 1 gram carbohydrate, except cottage cheese,
which contains more. GeneraUy speaking, where dairy products are
concerned, the lower the fat, the higher the sugar lactose, with skim
milkand "no fat"cheeses containingthe most lactoseand the least fat,
and butter containing no lactose and the most fat.
Neither butter nor magarine in myexperience wiU affect your blood
sugar significantly, and theyshouldn'tbea problem as far asweight is
concerned if you're not consuminga lot of carbohydrate along with
them. Margarine and most vegetable oUs contain trans fatty acids,
which are now considered unhealthy for the heart. Butter is now a
"healthy"fat. Organiccoconut oU isperhaps the healthiest and tastiest
oU for cooking and salads. Since it is soUd at room temperature, it
should be warmed sUghtly for salads. It can be found on the Internet
andin health food stores. Onetablespoon of cream hasonly0.5gram
carbohydrate —it would take 8 tablespoons to raise mybloodsugar
20 mg/dl.
Thecheese puffsI describe in the next chapter (page178)are lowin
carbohydrate and can be used instead of bread to make sandwiches.
Although personaUy I don't enjoyyogurt, many of my patients feel
theycannot survive without it. Forour purposesthe plainwholemilk
yogurt, unflavored, unsweetened, and without fruit, is a reasonable
food. A fuU 8-ounce container of plain, Erivan brand, unflavored
whole milk yogurt contains only 11 grams of carbohydrate and 2
156 Treatment
ounces of protein. You can even throw in some chopped vegetables
and not exceed the 12 grams of carbohydrate limit we suggest for
lunch. Do not use nonfat yogurt. The carbohydrate goes up to 17
grams per 8-ounce container.Yogurt can be flavored with cinnamon,
with DaVincibrand sugar-free syrups,with bakingflavor extracts, or
with the powder from sugar-freeJeU-0 brand gelatin (if the package
doesn't list maltodextrin as an ingredient) without affecting the car
bohydratecontent. It canbesweetened withstevia Uquid or powderor
with Equal or Splenda tablets that have been dissolved in a smaU
amount of hot water. Erivan brand yogurt is avaUable at health food
stores throughout the United States. If you read labels, you may find
brands simUarly low in carbohydrate in your supermarket; two such
brands are Stonyfield Farmand BrownCowFarm. Always be sure to
use only the whole-milkand not the low-fat products.
There are many soy products that can be used in our diet plan, and
soymilkis no exception. It's a satisfactory Ughtener for coffee and tea,
and one of my patients adds a smaU amount to diet sodas. Others
drink it as a beverage, either straight or with added flavoring such as
those mentioned for yogurt. PersonaUy, I find the taste too bland to
drink without flavoring, and I much prefer cream dUutedwith water.
When used in smaU amounts (up to 2 tablespoons/1 ounce), soymilk
need not be figuredinto the meal plan. It wiU curdle if you put it into
very hot drinks.
As noted in the No-No foods section, of the many brands of
soymilkon the market,WestSoy offers the onlyunsweetened ones I've
been ableto find, althoughother unsweetened brands are avaUable in
various parts of the country.
Soybean Flour
If you or someone in your home is wiUing to try bakingwith soybean
flour, youwUl find a neat solutionto the pastryrestriction. One ounce
of full-fat soybean flour (about lA cup) contains about 7.5 grams of
slow-acting carbohydrate. You couldmakechicken pies,tuna pies,and
evensugar-free JeU-0 piesor pumpkin pies.Just remember to include
the carbohydrate and protein contents in your meal plan.
Soybean flour usuaUy must be blended with eggto form a batter
suitable for breads, cakes,and the like. Creating a blend that works re-
Diet Guidelines Essential to theTreatment ofAllDiabetics 157
quires either experience or experimentation. Some recipes using soy
flour appear in Part Three, "YourDiabeticCookbook."
Bran Crackers
Of the dozens of different crackers that I have seen in health food
storesand supermarkets, I havefound onlythree brands that are truly
lowin carbohydrate.
• G/GScandinavian BranCrispbread, produced by G. Gundersen
Larvik A/S, Larvik, Norway (distributed in the United States by
Cel-Ent, Inc., Box 1173, Beaufort, SC29901, phone [843] 525-
1437). Each9-gramslice containsabout 3gramsof digestible car
bohydrate.If this product is not avaUable locaUy, youcan order it
directly from the importer. One case contains thirty 4-ounce
packages. They are also avaUable from Rosedale Pharmacy,
(888) 796-3348.
• Bran-a-Crisp, produced by SaetreA/S, N1411, Kolbotn, Norway
(distributed in the UnitedStates by Interbrands, Inc., 3300N.E.
164thStreet, FF3, Ridgefield, WA 98642). Each8.3-gramcracker
contains about 4 grams of digestible carbohydrate. Bran-a-Crisp
may be ordered directly from Interbrands, Inc., by phone or
e-maU if you cannot find it locaUy. Phone: (843) 524-9444;
e-maU: sales@branacrisp.com; Web site: www.branacrisp.com; or
order from Rosedale Pharmacy.
• Wasa Fiber Rye. These crackers are avaUable in most supermar
kets in the United States and in some other countries. One
crackercontains about 5gramsof digestible carbohydrate. Many
of mypatients feel that this is the tastiest of these three products.
Do not use other Wasa products, as they contain more carbohy
Although some people eat these without a spread, to me theytaste
like cardboard. My preference is to enjoy them with chive-flavored
creamcheese or butter. Crumbling two G/Gcrispbreads into a bowl
and coveringthem with creamor creamdiluted with water can create
bran crackercereal. AddsomeEqual or Splenda tablets (dissolved in a
bit of hot water) or someUquid stevia sweetener and perhapsa baking
flavor extract (banana flavor, butter flavor, et cetera), or one of the Da
Vinci sugar-freesyrups.
If eaten in excessive amounts, bran crackers can cause diarrhea.
158 Treatment
They should be eaten with Uquid. They are not recommended for
peoplewith gastroparesis (delayed stomach-emptying),sincethe bran
fibers can form a plug that blocksthe outlet of the stomach. The car
bohydrate in thesecrackers isveryslowto raiseblood sugar.They are
greatfor people whoneeda substitute for toast at breakfast.
Note: In the UnitedStates, labeling regulations requirethat fiberbe
Usted as carbohydrate. Therearemanydifferent kinds of fiber, soluble
and insoluble, digestible and undigestible, and so, because there is no
requirement to distinguish in labeling between them, these listings
can complicate computation of carbohydrate content. Usethe carbo
hydrate amounts that I have listed above instead of thoselisted on the
Toasted Nori
When my friend Kanji sent me a beautifuUy decorated canister from
Japan, I was mostimpressed andintrigued. You canimagine mydismay
when I removedthe coverand found seaweed. Mydismaywasonly tem
porary, however. I reluctandy opened one of the ceUophane envelopes
and puUed out a tissue-thin sUce. Myfirst nibblewasquitea surprise—
it was deUcious. When consumed in smaU amounts, I found, it had vir
tuaUy no effect upon bloodsugar. Onceaddicted, I combedthe health
food stores searching for more. Most of the seaweed I tried tastedlike
salty paper. EventuaUy, a patient explained to methat Kanji's seaweed is
a special kind caUed toasted nori. It contains smaU amounts of addi
tional ingredients that include soybeans, rice, barley, and red pepper. It
is avaUable at most health food stores, and is a very tasty snack. Fiveor
sixpieces at a timehave hadnoeffect uponmybloodsugar. TheClinis
tix/Diastix test showed no glucose after chewing. AstandardsUce usu
aUy measures VA x V/i inches and weighs about 0.3 gram. Since the
product contains about 40 percent carbohydrate, each strip wiU have
only 0.12 gramcarbohydrate. Larger sheets of toasted nori should be
weighed in order to estimate their carbohydrate content.
Sweeteners: Saccharin, Aspartame, Stevia, Splenda,
and Cyclamate
I carrya package of Equal (aspartame) tablets with me, particularly
whenI goout to eat. Cyclamate isnot currentlyavaUable inthe United
States, but maybe returning. Aspartame is destroyed by cooking and
is much more costly than saccharin, which has a bitter aftertaste, but
it wiU work for sweetening hot coffee or tea. I find that using one
Diet Guidelines Essential to theTreatment ofAll Diabetics 159
Vi-grain saccharin tablet for every Equal tablet rather than two sac
charin tablets or two Equal tablets eliminates saccharin's aftertaste and
keeps costs down. Equal tablets are avaUable in most pharmacies and
many supermarkets. Although Equal tablets contain lactose, the
amount is too small to affectblood sugar.
Acesulfame-K is a newartificial sweetener beingmarketedin tablet
form outside the United States by Hoechst, AG, of Germany. It is not
degraded by cooking. It is added to some "sugar-free" foods in the
United States under the brand name Sunett, and is combined with
glucose in the packaged powder caUed TheSweet One,whichyou ob
viously should avoid. There are, however, some questions about its
causingcancer, so there maybe better choices. Other noncaloric tablet
sweeteners wiUbe appearingon groceryshelves in the United Statesin
the future. Stevia, mentioned earUer, is an herbal sweetener and has
beenavaUable in healthfoodstores for manyyears. It is not degraded
by cookingand is packaged in tablet, powder, and Uquid forms. The
Uquid must be refrigerated to prevent spoiling. Stevia has not yet been
approved in the European Union because of fears that it maycause
cancer. Studies of this"possibUity" areunder way.
Splenda (sucralose) tablets are avaUable nowin some parts of the
UnitedStates, overseas, and on the Internet. Theyare benign in spite
of containing minuteamountsof lactose. In powdered form, Splenda,
like the others except stevia, isprincipaUy a mixture of sugars to pro
vide bulk and should be avoided.
No-Cal Brand Syrups
These artificiaUy sweetened liquidflavors aresoldbysomesupermar
kets in the northeastUnited States. (They aredistributed byCadbury
Beverages, Inc., Stamford, CT 06905-0800.) The avaUable flavors in
cludestrawberry, raspberry, black cherry, chocolate, and pancake/waf
fle topping. This product contains no calories, no carbohydrate, no
protein, and no fat. It takes a bit of imagination to put it to good use.
Forexample, I usedto spike mycoffee withthechocolate flavor, or my
teawithfruit flavors. I put the pancake/waffle topping on myeggs in
themorningafterheating it inaskiUet. Inrecent years, however, to my
taste, the chocolate flavor has deteriorated, soI no longeruse it.
Da Vinci Gourmet Sugar-Free Syrups
SimUar in concept to No-Cal syrups but in myopinionmuch tastier,
this product is avaUable from several Web distributors, including
160 Treatment
www.davincigourmet.com, and from Rosedale Pharmacy. Da Vinci
currentlyproduces morethanfortyflavors. Internetprices rangefrom
$7.49 to $8.95 for a 750 ml bottle. Flavorsinclude banana, blueberry,
caramel, cherry, chocolate, coconut, cookie dough, pancake, peanut
butter, and watermelon. I like to sometimes mix the toasted marsh-
maUow syrup into my morning omelet. For a list of distributors,
phoneDaVinci Gourmet, Ltd., at (800) 640-6779. Theproductiscer
tifiedkosher. Da Vinci alsoseUs syrups that are not sugar-free, so be
sure to specify sugar-free whenordering.
Water Sensations
Sweetened with sucralose and avaUable in several fruit flavors, this
flavor concentrate is sold in boxesof foU packets,which makes it ideal
for travel. It can be used to flavor water, seltzer, and yogurt and can
be added to tequUa to make a sugar-free margarita. For detaUs visit
www.watersensations.com or www.rx4betterhealth.com.
Flavor Extracts
There are numerous flavor extracts often used in baking that you can
use to make your food more exciting. They usuaUy can be found in
smaUbrown bottles in the bakingsupplyaisles of supermarkets. Read
carbohydrate content from the label. UsuaUy it's zero and therefore
won't affectyour blood sugar.
Mustard, Pepper, Salt, Spices, Herbs
Most commercial mustards are made without sugar and contain es-
sentiaUy no carbohydrate. This canreadUy be determined for a given
brand by reading the label or byusing the Clinistix/Diastix test. Pep
per andsalt have no effect uponblood sugar. Hypertensive individu
als with proven salt sensitivity should, of course, avoid salt andhighly
salted foods (seepage454).
Mostherbs and spices have verylowcarbohydrate content and are
usedin suchsmaU amounts that the amount of ingested carbohydrate
wiU be insignificant. Watch out, however, for certain combinations
such as powdered cinnamon with sugar. Just read the labels. By the
way, I mix powdered cinnamon with powdered stevia and cream
cheese and eat it off the platewithbites of smokedsalmon.
Low-Carbohydrate Salad Dressings
Most salad dressings areloaded withsugars and other carbohydrates.
The ideal dressing for someone who desires normal blood sugars
DietGuidelines Essential tothe Treatment ofAllDiabetics 161
would therefore be oU and vinegar, perhaps withadded spices, mus
tard, and foUowed by grated cheese or even real or soybacon bits.
There arenowavaUable some commercial salad dressings withonly1
gram carbohydrate per 2-tablespoon serving. This islowenough that
such a product can be worked into our meal plans. Be careful with
mayonnaise. Most brands are labeled "carbohydrate —0 grams," but
may contain up to 0.4 grams per tablespoon. This is not a lot, but it
adds up if you eat large amounts. Some imitation mayonnaise prod
ucts have 5grams of carbohydrate per2-tablespoon serving. I person
aUy use coconut oU and vinegar on mysalads, but I like to mix the
vinegar withDaVinci sugar-free raspberry syrup.
Although aU nuts contain carbohydrate (as weU as protein and fat),
they usuaUy raise blood sugar slowly and can in smaU amounts be
worked into meal plans. As withmost other foods, youwiU want to
look up your favorite nuts in oneof the books listed in Chapter 3 in
order toobtain their carbohydrate content. Byway ofexample, 10 pis
tachio nuts (smaU, not jumbo) contain only 1 gram carbohydrate,
whUe 10 cashewnuts contain 5grams ofcarbohydrate. Although afew
nuts may contain little carbohydrate, the catch is in the word "few."
Very few ofuscan eatonly afew nuts. Infact, I don'thave asingle pa
tient who can count out a preplanned number of nuts, eat them, and
then stop. So unless you have unusual wiUpower, beware. Just avoid
themaltogether. Also beware of peanut butter, another deceptive ad
diction. One tablespoon of natural, unsweetened peanut butter con
tains 3grams ofcarbohydrate, andwiU raise mybloodsugar 15 mg/dl.
Imagine theeffect onblood sugar of downing 10 tablespoons.
Sugar-Free JeU-0 Brand Gelatin
This is one of the few foods that in smaU amounts wiUhave no effect
upon bloodsugar if youget the kindthat is indeed sugar-free. I have
found it tobetruethat inmy area "sugar-free" actuaUy contains some
maltodextrin, which is a mixture of sugars andwiU raise yourblood
sugar. The ready-to-eat variety in plastic cups does not thus far con
tain maltodextrin —or at least thatwhich I've found onmygrocery's
shelves. Check thelabels. Truly sugar-free JeU-O or other truly sugar-
free brands ofgelatin are fine for snacks anddesserts. AVi-cup serving
contains no carbohydrate, no fat, and only1gramof protein. Just re
member not to eat so much that you feel stuffed (see "The Chinese
162 Treatment
Restaurant Effect," in Chapter 6). You can enhance the taste by pour
ing a littleheavycreamoveryour portion. One of mypatients discov
ered that it becomes eventastier if you whip it in a blender with cream
whenit has cooled, just beforeit sets. Of the manyflavors of sugar-free
JeU-0 that are avaUable, I likeapple, Hawananpineapple, and water
melon. Unfortunately, very fewsupermarkets seemto carry the apple
and Hawanan pineapple flavors, and I wonder if they stiU exist.
If the only"sugar-free" JeU-0 you can find contains maltodextrin,
try adding some Uquid stevia and Da Vinci sugar-free syrup to Knox
unflavoredgelatinas a tastysubstitute.
Sugar-Free Jell-0 Puddings
Available in chocolate, vaniUa, pistachio, and butterscotch flavors,
these make a nice dessert treat. Unlike JeU-0 gelatin, they contain a
smaU amount of carbohydrate (about 6 grams per serving), which
should be counted in your meal plan. Instead of mixing the powder
with milk, use water or water plus cream. Every 2 tablespoons of
creamwUl add 1gramof carbohydrate.
Chewing Gum
Gum chewing can be a good substitute for snacking and can be of
value to people with gastroparesis becauseit stimulates salivation and
saUva contains substances that facUitate stomach-emptying. The car
bohydrate content of one stickof chewing gum varies from about 1
gramin astickof sugar-free Trident or Orbit (tastes better) to about 7
grams per piece for some Uquid-fiUed chewing gums. The 7-gram
gumwiU rapidly raise mybloodsugarbyabout 35mg/dl. The carbo
hydrate content of a stickof chewing gum can usuaUy be found on
the package label. "Sugar-free" gums aU contain smaU amounts of
sugar — the primaryingredientof Trident"sugarless" gumissorbitol,
a corn-basedsugaralcohol. It alsoincludes mannitol and aspartame. I
sometimes use a chewing gum called XlearDent. It contains 0.72
grams of thesugarxylitol per piece. XyUtol isanantimetabolite (meta
bolicpoison) for bacteria andprevents tooth decay whenchewed reg
ularly. It may be obtained by phoning (877) 599-5327 or on the
Internet at www.xlear.com.
Very Low Carbohydrate Desserts
Part Three of this book consistsof low-carbohydrate recipes, prepared
and testedbychefs. It includeseasyrecipes for somelow-carbohydrate
Diet Guidelines Essential to theTreatment ofAll Diabetics 163
desserts that are truly delicious. Morelow-carb desserts can be found
in my book The Diabetes Diet, Little, Brown, 2005.
Coffee, Tea, Seltzer, Mineral Water, Club Soda, Diet Sodas
None of these products should have significant effect upon blood
sugar. The coffee and tea may be sweetened with Uquid or powdered
stevia, or with tablet sweeteners such as saccharin, cyclamate, su
cralose (Splendatablets),steviatablets, and aspartame (Equaltablets).
Remember to avoid the useof more than 2 teaspoons of cow's milkas
aUghtener. Tryto usecream (which hasmuchless carbohydrate, tastes
better,and goesmuch further). Readthe labels of "diet"sodas, as a few
brands containsugar in the formof fruit juices. Manyflavored mineral
waters, bottled "diet" teas, and seltzers also contain added carbohy
drateor sugar, as do manypowdered beverages. Again, readthe labels.
You can also try adding your flavor choiceof Water Sensations or Da
Vinci sugar-free syrupsto seltzer to create your owndiet soda.
Frozen Diet Soda Pops
Many supermarkets and toy stores in the United States seU plastic
moldsfor making your ownicepops.If these arefilled withsugar-free
sodas, youcancreate a tastysnackthat hasnoeffect uponbloodsugar.
Do not usethe commerciaUy made"sugar-free" or "diet"icepops that
are displayed in supermarket freezers. They contain fruit juices and
other sources of carbohydrate.
Alcohol, in Limited Amounts
Ethyl alcohol (distiUed spirits),aswediscussed on pages 136-137, has
no direct effect upon blood sugar. Moderate amounts, however, can
have a rapideffect upon theUver, preventing the conversion of dietary
proteinto glucose. If youarefoUowing a regimen that includes insulin
or a pancreas-stimulating oral hypoglycemic agent, you'redependent
upon conversion of protein to glucose in order to maintain blood
sugar at safe levels. The effects of small amounts of alcohol (i.e., Wi
ouncesof spirits for a typical adult) are usuaUy negligible. Most con
ventional American beers (Ught lagers), in spiteof their carbohydrate
content, don't seemto affect blood sugar when only one can or bottle
is consumed. Darker beers, such as ales, stouts, and porters, can con
tain considerably more carbohydrate, and since beer does not have
Nutrition Factslabeling, findingthe true carbohydrate content can be
difficult. "Lite"beerswiU generaUy have the least carbohydrate.
164 Treatment
Read Labels
VirtuaUy aU packaged foods bear labels that reveal something about
the contents. The FDA nowrequiresthat labelsof packagedfoods list
the amount of carbohydrate, protein, fat, and fiber in a serving. Be
sure, however, to note the sizeof the "serving." Sometimes the serving
sizeis so smaU that you wouldn't want to be bothered eating it.
Beware of labelsthat say"Ute," "Ught,""sugar-free,""dietetic," "diet,"
"reduced-calorie,""low-calorie," et cetera. Counts of calories are only
going to teU you so much, and "lowfat" is going to teU you nothing
about carbohydratecontent."Fat-free" desserts maybe the most dan
gerous of aU. Even ifyou're losing weight, carbohydrate intakewiU im
pede your efforts much more than fat wiU (see Chapter 9). For
example,I've found that it's impossible to put weight on veryslimpa
tients foUowing low-carbohydrate dietsbygivingthem 900extra calo
ries a day in the form of 4 ounces of olive oU. Two recent studies
support this — but onlyif carbohydrate is verylimited. They showed
that when carbohydrate is low, the fat is metaboUzed, not stored.
"Low-fat"and "fat-free"foods frequentlybut not always contain more
carbohydratethan the foodstheyreplace. The onlywayyou can deter
mine the carbohydrate content is to read the amount stated on the la
bel. But even this can be deceptive. For example, one popular brand of
"sugar-free" strawberry preserves has a label that states, "Carbohy
drate — 0."Yet anyone can see the strawberries in the jar, and com
mon sensewould teU you that strawberries contain carbohydrate. So
deceptive labeling occurs and, in myexperience, is fairly prevalent in
the "diet" food industry.
Use Food Value Manuals
In Chapter 3, severalbooks are listedthat showthe carbohydrate con
tents of various foods. These manuals are recommended but not es
sential tools for creatingyour meal plan. The guidelinesand adviceset
forth in Chapters 9-11 of this book, plus perhaps the recipes in Part
Three and in TheDiabetes Diet, are aU you reaUy need to get started.
If you want the potential for considerable varietyin your meals,get
aU the books Ustedin Chapter 3. The best of these is TheNutriBase
Complete Book ofFood Counts. Food Values ofPortions Commonly Used
Diet Guidelines Essential to the Treatment ofAll Diabetics 165
has been the dietitian's bible for more than fifty years. It is updated
everyfewyears. Besure to use the indexat the back to locate the foods
of interest. Notethat on every page in the mainsection, carbohydrate
and fat content are listedin the samecolumn. The carbohydrate con
tent of a foodalways appears belowthe fat content.Donot get the two
confused. Also, be sure to note the portion sizein aU these books.
The USDA's nutrient database is a very handy resource and of
fers software for use on PCs and handheld PDA computers that wiU
enableyou to find nutrition informationon just about any food you
choose. The Nutrient Data Laboratory home page can be found at
It is common practice to prescribe supplementary vitaminsand min
erals for diabetics. This is primarily because most diabetics have
chronicaUy high blood sugars and therefore urinate a lot. Excessive
urinationcauses a loss of water-soluble vitamins and minerals. If you
cankeep your bloodsugars lowenough to avoid spilling glucose into
the urine (you can test it with Clinistix/Diastix), and if you eat red
meat at least once or twice a week, and a variety of vegetables, you
should not require supplements. Note, however, that major dietary
sources of B-complex vitamins include "fortified" or supplemented
breads and grains in the United States. If you're foUowing a low-
carbohydrate diet and therefore exclude these fromyour meal plan,
youshouldeat somebeansprouts,spinach, broccoli, brussels sprouts,
or cauUflower eachday. If youdo not like vegetables, youmight takea
B-complex capsule or a multivitamin/mineral capsule each day. See
page 179 for a discussion of calcium supplementation for certain
peoplewho foUow high-fiber or high-protein dietsor usemetformin.
Supplemental vitamins and minerals shouldnot ordinarUybe used
in excess of the FDA's recommended daUy requirements. Large doses
can inhibit the body's synthesis of some vitamins and intestinal ab
sorption of certain minerals. Large doses are also potentiaUy toxic.
Doses of vitamin C in excess of 500 mg daUy may interfere with the
chemical reaction on your blood sugar strips. Asa result, your blood
sugar readings can appear erroneously low. Large doses of vitamin C
can actuaUy raiseblood sugar and evenimpair nervefunction (as can
doses of vitamin B-6in excess of 200mg daUy). Vitamin E has been
166 Treatment
shown to reduce one of the destructive effects of high blood sugars
(glycosylation of the body's proteins), in a dose-dependent fashion—
up to 1,200 IU (international units) per day. It has recently been
shown to lower insulin resistance. I therefore recommend 400-1,200
IUper dayto a number of mypatients. Besure to use the forms of vi
tamin E known as gamma tocopherol or mixed tocopherols, not the
commonalphatocopherol, whichcaninhibit the absorptionof essen
tialgamma tocopherol from foods andat high doses hasactuaUy been
shown to increase risk of cardiac death.
Anewdiet oftenbringsabout changes in frequency and consistencyof
bowel movements. This is perfectly natural and should not cause con
cernunless youexperience discomfort. Increasing the fiber content of
meals, as with salads, bran crackers, and soybean products, can cause
softer and more frequent stools. More dietary protein can cause less
frequent and harder stools. Calcium tablets cancause hard stools and
constipation, but this is usuaUy offset if they contain magnesium.
Normal frequency of bowel movements can range from 3 times per
dayto 3timesper week. Ifyounotice anychanges inyourbowel habits
more or lessthan these frequencies, discussthem with your physician.
Most of mypatientsinitiaUy feel somewhat deprived, but alsograteful
because they feel more alert and healthier. (See the chapter "Before
and After" for reactions of some patients to the new diet.) I faU into
this category myself. Mymouthwaters whenever I pass a bakeryshop
and sniff the aroma of fresh bread, but I am alsograteful simplyto be
aliveand sniffing.
Creating a Customized Meal Plan
Nowthat you have the essentials of what you should eat and
what youshould avoid, it'stime totake you through thesteps
of customizing a meal plan that wiU getyouon yourway to
blood sugar normalization.
If youfound yourself thinking asyou went through the No-No foods
section of the prior chapter that aU of this information goes against
conventional thinking —you're right. Nodoubtasyouembark upona
meal and treatment planto normalize your bloodsugars, weU-mean-
ing but Ul-informed friends and relatives wiU urge you to try more
"fun" foods, or toeat less fat andmore "complex" carbohydrates. I sug
gest that youread AppendixA, which provides some possible explana
tionsasto whyconventional wisdom mayhave taken a wrongturn.
If you usebloodsugar-lowering medications suchas insulin or oral
agents, the first ruleof meal planning isdon't change your diet unless
yourphysician first reviews thenewmeal planandreduces your med
ications accordingly. Most diabetics who begin our low-carbohydrate
diet showanimmediate anddramatic drop inpostprandial blood sugar
168 Treatment
levels, as compared to blood sugars on their prior, high-carbohydrate
diets. If at thesametimeyour medications are not appropriatelyre
duced, your bloodsugars candrop to dangerously lowlevels.
The initial meal planshouldbe geared toward blood sugar control,
andalso toward keeping youcontentwithwhat you eat. Sowith those
thingsin mind, if I were to sit downwith youto"negotiate" yourmeal
plan, I wouldneedto have before me aGlucograf data sheet (Chap
ter 5) showing blood sugar profiles and blood sugar-lowering med
ications (if any) takenduring the preceding week. I also wouldaskfor
a Ust of what and when you eat on a typical day. This information
wouldgive me anidea of what youlike to eat andwhateffect particu
lar doses of blood sugar-lowering medications have on your blood
sugars. I also must know your current weight and about any other
factors — such as delayed stomach-emptying and medications for
other ailments — that might affect your blood sugar. In negotiating
the meal plan, I'dtry wherever possible to incorporate foods you like.
We wiU discuss weight reduction in Chapter 12. Changes for this
purpose can bemadeafter observing theeffects of the initial diet for a
month or so.
If you've trieddieting to lose weight orto control yourbloodsugar,
you may have foundthat simplycutting backon calories according to
preprinted tables or fixed calculations can be frustrating andcaneven
havethe opposite effect. Say you have a supperthat'stoo smaU to sat
isfy you. Later you're so hungryyou feel you must have a snack. If
you're likemost people, yoursnack wUl likely be snack food, abowl of
cereal, or some fruit — that is, somethingloadedwith carbohydrate—
so you end up with high blood sugars and more calories than you
would haveconsumed if you'd started with a sensible meal. My expe
rience is that it's always best to start with a plan that aUows you to get
up fromthe table feeUng comfortable but not stuffed.
If you've ever foUowed theoldADA"exchange" system for preparing
diabetic mealplans, you'U find that keeping track of grams of carbohy
drate and ounces of protein food (we always estimate carbohydrate in
grams and protein foods in ounces) requires considerably less effort
Not onlyisit easier thantheexchange approach, it's moreeffective, be
cause it places the focus onthenutrientsthat actuaUy affect bloodsugar.
Since aU of my patients bringme glucose profiles, overthe years it has
not been very difficult to develop guidelines for carbohydrate con
sumptionthat makebloodsugar control relatively easywithout causing
too great a feeling of deprivation, even for those tryingto lose weight.
Creatinga CustomizedMealPlan 169
Mybasicapproachin negotiating a meal plan is that I first set car
bohydrate amountsfor eachmeal. ThenI askmypatient to teUmehow
many ounces of protein we should add to make him/her feel satisfied.
(I actuaUy showthem plasticsamplesof protein foods of various sizes,
to help themestimate amounts.) Forexample, I usuaUy advise patients
to restrict their carbohydrate intake to no morethan 6 gramsof slow-
acting carbohydrate at breakfast, 12 grams at lunch, and 12 grams at
supper.* FewpeoplewouldbewiUing to eat less than theseamounts of
carbohydrate. (These guidelines also apply to chUdren.) There is no
such thingas an essential carbohydrate for normal development, de
spitewhat the popular press might have you beUeve, but there most
certainly are essential amino acids (protein) and essential fattyacids.
As mentionedin Chapter 9, the main reason I don't suggest that you
avoid aU carbohydrate is that there are many constituents of vegeta
bles —suchas vitamins and minerals, but also manyother nonvita-
min chemicals (phytochemicals) —that are only recently becoming
understood but that are nonetheless crucial to diet and cannot be ob
tainedthroughconventional vitamin supplements. Thisisparticularly
true for whole-plant and leaf varieties. FoUc acid — so-caUed because
it isderived fromfoUage —isessential to aU mannerof development,
but strictlyspeakingis neither vitaminnor mineral.
IdeaUy, your blood sugar should be thesame after eating as it was
before. If blood sugar increases bymore than 10 mg/dl after a meal,
evenif it eventuaUy drops to your targetvalue, either the meal content
should be changed or blood sugar-lowering medications should be
usedbefore you eat. Contraryto ADA guidelines, it has recently been
shown that postprandial, or after-meal, bloodsugars are more likely
than fasting bloodsugars to cause cardiovascular damage.
Distinctions are often made between "complex" and "simple" carbo
hydrates, with foods suchas multigrain breads or pastatouted as"fuU
of complex carbohydrates." This is essentiaUy a meaningless distinc-
*It'snot at aU necessary to consume carbohydrate of any type for breakfast. If
youdo,theonlykindI recommend isintheform ofacceptable vegetables (which
can work well in, for example, an omelet) or the bran crackersmentioned in the
previous chapter.
170 Treatment
tion, if not a fooUsh one. There are fast-acting carbohydrates —
starches and sugars that break down rapidly and have a consequent
rapid effect on blood sugars — and there are slow-acting carbohy
drates. GeneraUy, slow-acting carbohydrate comes from whole-plant
vegetables (andothers listed on page 151). Theyare predominantly in
digestible fiber accompanied by some smaU amount of digestible car
bohydrate and vitamins, minerals, and other compounds, but have
relatively Uttle effect on blood sugars.
The foods in the foUowing Ust areslow-actingcarbohydrate foods.
These can constitute the buUdingblocks of the carbohydrate portion
of eachmeal. Of courseyou needn't limit your foods to these — many
other suchbuUding blocks can be created. Read labels on packaged
foods, consult nutrition tables for carbohydrate values of foods you
like, checkyourbloodsugars, and find out which foods work for you.
Equivalent in blood sugar effect to approximately 6grams of
carbohydrateper serving
• 6 Worthington Stripples or Morningstar Farms Breakfast Strips
(meatless soybacon) (also contains 1ounce protein)
• 3 Morningstar Farms Breakfast Links (meatless soy sausage)
(also contains 2 ounces protein)
• Vh Bran-a-Crisp crackers
• 2 G/G crispbreads
• 1WasaFiber Ryecracker
• 4!/2 ounces Brown CowFarmor Stonyfield Farmwhole-milk un
flavored yogurt (8ounces contains 11 grams carbohydrate and 1
ounce protein)
• 1cup mixedsalad with oU-and-vinegar (not balsamic) dressing
• % cup cooked whole slow-acting carbohydrate vegetable (or lA
cup mashed or Vi cup slicedor diced) from list on page 151
• 1 serving Jett-0 sugar-free pudding made with water or water
and 1tablespoon cream
• Vi medium avocado (3 ounces)
Equivalent inblood sugar effect toapproximately 12grams of
carbohydrate per serving
• 1cup mixed salad with oU-and-vinegar (not balsamic) dressing,
plus%cup cooked wholevegetable (or V4 cup mashedor Vi cup
sliced or diced) from list on page 151
Creatinga CustomizedMealPlan 171
• 1 cup mixed salad prepared with 4 tablespoons packaged dress
ing (if each tablespoon contains 1.5 grams of carbohydrate)
• 8 ounces BrownCowFarm or StonyfieldFarm whole milk unfla
vored yogurt (contains 11 grams of carbohydrate plus 2 ounces
TheseUsts sUghtly exaggerate the carbohydrate content of salad and
cooked vegetables, but because of their bulk and the Chinese restau
rant effect, the net effect upon blood sugaris approximately equiva
lent to the amounts of carbohydrate shown. To this slow-acting
carbohydrate, we'd add an amount of protein that, in your initial
opinion, would aUow you to leave the table feeling comfortable but
not stuffed.
Aswith carbohydrate, it isnecessaryto keepthe sizeof the protein por
tion at a particular meal constant from one day to the next, so if you
eat 6 ounces at lunch one day, you should have 6 ounces at lunch the
next. This is especiaUy important if you'retakingblood sugar-lowering
medications. If you're using tables of food values and need to convert
grams of protein to ounces of a protein food, keep in mind that for
these meal plans, 6 grams of protein is the equivalent of 1 ounce of
an uncooked protein food. To estimate by eye, a portion the size of a
deck of playingcards weighs about 3 ounces (red meats weigh about
3.7 ounces because of their greater density).
In order to maintain muscle mass, most people should consume at
least 1-1.2 gramsof protein per ldlogram of idealbody weight.*Ath
leteswitt requireconsiderably more, aswiU growing chUdren.
Proteinfoods with virtually no carbohydrate
• Beef, lamb, veal
• Chicken, turkey, duck
• Eggs
• Most cold cuts (bologna, salami, et cetera)
• Fish and shellfish (fresh or canned)
*This would amount to 11.5-14 ounces of protein daily for a nonathletic indi
vidual whose ideal body weight is 155pounds.
172 Treatment
• Most frankfurters
• Pork (ham, chops, bacon, et cetera)
• Most sausages
Proteinfoods with a small amount ofcarbohydrate(1gram
carbohydrateper ounceofprotein)
• Cheeses (other than cottagecheese and feta cheese); the gram of
carbohydrateper ouncefound in most cheeses should usuaUy be
included when computing the carbohydrate portion of a meal
Soyproducts (up to 6 gramscarbohydrate per ounceofprotein —
checknutrition label onpackage)
• Veggie burgers
• Tofu
• Meatless bacon
• Meatless sausage
• Other soysubstitutes (for fish, chicken, and so on)
If you have a rare disorder caUed famUial dyslipidemia, where di
etary fat actuaUy can increase LDL, restrictions on certain types of
dietary fats contained in some protein foods may be appropriate.
Mealsneed not foUow a rigidlyfixed time schedule, provided, in most
cases, that you do not begin eatingwithin 4 hours of the end of the
prior meal. This is so the effect of the first meal upon blood sugar
won't significantly overlap that of the next meal. For those who inject
insulin before meals, it's veryimportant that meals be separated by at
least 5hours if you want to correct elevated blood sugarsbeforeeating
(seeChapter 19).This isalsoideaUy but not always true of snackscov
ered by insulin.
Snacks are permitted for some diabetics but certainly not required.
The carbohydrate content of snacksmaydupUcate but should not ex
ceed that aUocated for lunch or supper. So if you ate lunch at noon,
you might tolerate a snack that didn't exceed 12 grams of carbohy
drate at about 4 p.m. You would then eat supper at about 8 p.m. Snacks
are discussed in greater detaU later in the chapter.
Creatinga CustomizedMealPlan 173
If you do not take insulin, you need not be restricted to only three
daUy meals if you prefer four or more low-carbohydrate meals on a
regularbasis. The timing, again, shouldideaUy be at least4 hours after
the end of the prior meal or snack. For most type 2 diabetics, it maybe
easier to control blood sugar,with or without medication, after eating
several smaUer mealsthan after eatingonlyone or twolargemeals.
Remember that there are no diabetes-relatedprohibitions on coffee
and tea, either plain or with limited cream (not milk) and/or tablet
(not powdered, except for stevia) sweeteners.*
Now, let's attempt to translate our guidelines into some practical
With or without blood sugar-lowering medications it is usuaUy more
difficult to prevent a blood sugar rise after breakfast than after other
meals. Therefore, for the reasons discussed under "The Dawn Phe
nomenon" in Chapter 6,1 usuaUy suggest half as much carbohydrate
at breakfast as at other meals. Your body wiU probably not respond
as weU to either the insulin it makes or to injected insulin for about
3 hours after you get up in the morning because of the dawn phe
It is wise to eat breakfast everyday, especiaUy if you're overweight.
In my experience, most obese people havea history of either skipping
or eating very httle breakfast. They then become hungry later in the
dayand overeat. Nevertheless, for most of us, anymeal can be skipped
without adverse outcomes, provided, of course, that insulin or any
other blood sugar-lowering medication taken specifically to cover
that meal is also skipped.
A typical breakfast on our meal plan would include up to 6 grams
carbohydrate and an amount of protein to be determined initiaUy by
you. There are numerous possible sources of appetizing ideas for the
carbohydrateportion of your breakfast. The best placeto start is with
what you currently eat, as longas it's not on the No-Nolist (seepages
152-153). You can also sample recipes from Part Three or from my
book The Diabetes Diet. You can experiment with foods in the "So
*Remember that 10cups of coffee, eachwith 2 tablespoons cream, can raise the
blood sugar of a 140-pound type 1diabeticby50 mg/dl.
174 Treatment
What's Left to Eat?" section, on pages 150-163. There are many soy
bean products, such as the foods mentioned on page 154. Despite re
strictions, with aUttle creativity you canfindanynumber of satisfying
things to have for breakfast.
Suppose that, like many of my newpatients, you'vebeen eatingfor
breakfast abagelloadedwith creamcheese and 2 cups of coffee with
skim milk and Sweet'n Low powdered sweetener (totaling about 40
grams of rapid-acting carbohydrates altogether). As we negotiate, I
might propose that you substitute other sweeteners for the Sweet'n
Low and 1 ounce WestSoy soymilk (0.5 gram carbohydrate) for the
skim milk in each cup of coffee (or use cream). Then I'd recommend
that instead of abagel you eat aBran-a-Crisp cracker (4 grams carbo
hydrate) with 1 ounce of cream cheese (1 gram carbohydrate plus 1
ounce protein). This adds up to about 6 grams of carbohydrate. Fi-
naUy, I'd suggest that youadda protein food to your mealto make up
forthe calories and"filling power" that disappeared with the bagel.
Let's sayyou decide you'U eat eggs for breakfast (or egg whites or
Egg Beaters, although for most of us on alow-carbohydrate regimen,
neither of theseis necessary for cholesterol control). I'daskhowmany
eggs it would take to make you feel satisfied after giving up the bagel.
You might want to makeavegetable omelet instead of eating one of
the carbohydrate foods mentioned above. If you're unnecessarUy
afraid of eggyolks, you might use organic eggs or eggwhites. If you
find egg whites bland, you couldadd spices, or soy or Tabasco sauce,
or some mushrooms or a smaU amount of onion or cheese, or chUi
powder, or even cinnamon with stevia, to enhance the taste. One of
my current personal favorites for flavoring isa"chUi sauce" made from
Better Than BouiUon ChUi Base. This packs a nice chUi punch with
veryUttle carbohydrate (according to thelabel, 1gramof carbohydrate
per 2 teaspoons), and works quite weU on eggs or other foods, de
pendingon your tastes. (I liketo make chUi burgers with it — which
youcouldcertainly have for breakfast.) This product, akind of mushy
paste, comes in a smaU glass jar at most supermarkets or from Supe
rior QuaUty Foods, 2355 E. Francis St., Ontario, CA 91761 (e-maU:
service@superiortouch.com; ontheWebatwww.superiortouch.com).
Some years ago, I triedto helpmy patients who felt they hadto have
cold cereal include a smaU amount in their breakfast meal plan, but
blood glucose profiles showed consistently that this just didn't work.
Grain products, with the exception of the bran products I've men
tioned, contain too much fast-acting carbohydrateto aUow us to keep
Creatinga CustomizedMealPlan 175
blood sugarunder control, and sowe'vehad to eliminate breakfast ce
realsentirely. An alternativemight be the bran crackercerealdescribed
on page 157.
The good news is that there are lots of other tasty, filling things
to eat.
If you don't want eggs, you might try some smoked fish, tuna fish,
or even a hamburger. I have one patient who eats two hot dogs for
breakfast — her favorite food. The quantity of fish or hamburger
would be up to you, but it would haveto be kept constant from one
day to the next. You can either weigh the protein portion on a food
scale or estimateit by eye.The rule of thumb is, again, that a portion
of poultry or fish the size of a standard deck of playing cards wiU
weigh about 3 ounces (3.7 ounces for redmeat). One egg hasthe ap
proximateproteincontent of 1ounceof meat, poultry,or fish plus up
to 0.6 gramcarbohydrate.
You cantakeany of the foods in the 6 grams of carbohydrate Ust on
page 170and add protein to them (cheese, eggs, et cetera) to make a
satisfying breakfast. You can haveless than 6 grams carbohydrate or
even no carbohydrate, providedthe amount is unchanged from day
to day.
FoUow the same guidelines for lunch as for breakfast, with the excep
tion that the carbohydrate content may be doubled, up to 12grams.
Say, forexample,that you andyour friends goto lunch everydayat
the "greasy spoon"around the corner fromwork and areservedonly
sandwiches. You might try discarding the sUces of bread andeating the
filling — meat, turkey, cheese, or other protein food— with a knife
and fork. (If you choose cheese, remember to count 1gram carbohy
drate per ounce.) You could alsoorder ahamburger without the bun.
And instead of ketchup, you could use mustard, soy sauce, or other
carbohydrate-free condiments. You then might add Mb cups cooked
wholevegetable fromthe Ust on page 151 (12grams carbohydrate) or
2 cups of salad with vinegar-and-oU dressing (12 grams of carbohy
drate) to round out your meal.
If you want to create a lunch menu from scratch, use your food
valuebooks to look up foods that interestyou. If you like sandwiches,
one double cheese puff as described later in this chapter under the
176 Treatment
"Snacks" heading wiU be about the size of a large slice of bread.
They're sturdy enough that you could make a sandwich from two of
them. Just make sure to account for the protein and carbohydrate in
the cheese.
The foUowing buUding blocks may be helpful in giving you a
Forthe protein portion, oneofthefollowing
• A smaU can of tuna fish contains 3V4 ounces by weight in the
United States. If you're packing your lunch, these can be quite
convenient if you like tuna. The next larger size can contains 6
ounces. The tastiest cannedtuna I've tried is made by Progresso,
packedin oliveoU.
• 4 standard sUces of packaged pasteurized process American
cheese (process cheddar in the U.K.) weigh about 2% ounces.
This wUl contain about 3 ounces of protein and 3 grams of car
Forabout 12gramscarbohydrate, oneofthefollowing
• VA cups whole cookedvegetables (from the list on page 151).
• 2 cups mixed green salad, with 1 sUce of tomato and vinegar-
and-oil dressing. Sprinkling bacon or soy bacon bits or grated
cheesewill havenegligible additionalblood sugareffect.
• V/i cups salad, as above, but with 3 tablespoons of commercial
salad dressing (other than simple vinegar-and-oU) containing 1
gram of carbohydrateper tablespoon. Check the label.
Youmight decide that 2cupsof salad withvinegar-and-oU dressing
is fine forthe carbohydrate portion of your lunch. You then should de
cide how much protein must be added to keep you satisfied. One per
son might be happy with a 3xA-o\xnce can of tuna fish, but another
might require 2 large chicken drumsticks or a packet of lunch meat
weighing 6 ounces. For dessert, you might want some cheese (in the
European tradition) or perhaps some sugar-free JeU-O gelatin (if it
contains no maltodextrin) coveredwith 2tablespoons ofheavy cream.
You might consider some of the dessertsdescribed in PartThree or in
The Diabetes Diet. The possible combinations are endless; just use
your food value books or readlabels for estimating protein and car
bohydrate. Some people, after having routinely eaten the same thing
for years, discoverthat their new meal plan opens up culinary possi-
Creatinga CustomizedMeal Plan 177
bUities theyneverknewexisted. Our patients,asweU as readers,are al
ways looking for recipes, so if you come up with a recipe that you
think is particularly good, pleasefeelfreeto share it on the Website for
this book. You can either go directly to www.diabetes-book.com/
recipes/recipes.shtml or choose the "Recipes" link and foUow the di
rectionsfor submission. The simplerecipe format used in Part Three
must be foUowed precisely for it to be acceptable.
Supper should foUow essentiaUy the sameapproach as lunch. There is,
however, one significant difference that wiU especiaUy apply to those
who are affected by delayed stomach-emptying (gastroparesis) and
take insulin. As we've discussed briefly, this condition can cause un
predictable shifts in blood sugar levels because food doesn't always
pass into the intestines at the same rate from meal to meal. The diffi
cultywith supper is that you can end up with unpredictably high or
lowblood sugars whUe you are sleeping and unable to monitor and
correct them. Sustained exposure to high blood sugars whUe
sleeping — evenif they are normalized during the day— can lead to
long-termdiabeticcompUcations. For certainaffected people, aviable
approachto this problemis to facUitate stomach-emptying byreplac
ing salads with cooked vegetables (from our list) that are lowin insol
uble fiber and reducingprotein content. For thesepeople, the amount
of protein at supper wouldbe less than that eatenat lunch — just the
opposite of what has become customary for most Americans. A more
completeanalysis of this problemappears in Chapter 22.
If you likecookedvegetables (fromour Ust) for supper, remember
that most can be interchanged with salads as near equivalents —
%cupof cookedwholevegetable (or lA cupmashedor V4 cup sUced or
diced) and 1 cup of salad each havethe blood sugar effect of about 6
grams carbohydrate.
If you like wine with dinner, choose a very dry variety and limit
yourself to one 3-ounce glass (see pages 136-137 for further detaUs).
Asnoted on page 163,one beer mayactuaUy turn out to have no effect
upon your blood sugar. Still, don't drink more than one if you take
insulin or use one of the oral agents that stimulate insulin secretion
(sulfonylureas; remember that our program prohibits use of these
178 Treatment
For many people with diabetes, snacks should be neither mandatory
nor forbidden. They do, however, pose a problem for people who take
fast-acting insulin before meals. Snacks should be a convenience, to
relieve hunger if mealsaredelayed or spacedtoo far apart for comfort.
If your diabetes is severe enough to warrant the use of rapid-acting
blood sugar-loweringmedicationbeforemeals, such medication may
also be necessarybefore snacks.
The carbohydrate limit of 6 grams during the first fewhours after
arising and 12grams of carbohydrate thereafter that appUes to meals
also appUes to snacks. Besure that your prior meal has been fuUy di
gestedbeforeyour snackstarts (this usuaUymeanswaiting4-5 hours).
This is so that the effects upon blood sugar wUl not add to one an
other.You needn't worry, however, if the snackissosparse(say, a bit of
toasted nori) as to have negUgible effects on blood sugar. Sugar-free
JeU-O gelatin (without maltodextrin) can be consumed pretty much
whenever you like, provided you don't stuff yourself and provoke the
Chinese restaurant effect. As a rule, snacks limited to smaU amounts of
protein wiU have less effect upon blood sugar than those containing
carbohydrate. Thus 2-3 ounces of cheese or cold cuts might be rea
sonable snacks for some people.
Among my patients, a common favorite snack,which has a negUgi
ble amount of carbohydrate, is homemade microwave cheese puffs.
They're simple and convenient to make. Get some freezer paper from
the grocery— not waxed paper. It has a duU side and a shiny side.
Place a sliceof American cheese (process cheddar in the U.K.) on the
shiny side of a piece of the freezer paper, then pop it into the
microwave for 1-2 minutes, depending on how powerful your mi
crowave is. The cheese wiU bubble up and puff quite nicely, but let it
cool a Uttle before attempting to remove it from the paper. CooUng
can be accelerated byputting it into the freezer for 30seconds.
Twoslicesside by side in the microwave wiU make a double cheese
puff. Two of these are suitable for a sandwich. I have put mayon
naise on one and mustard on the other and ham or turkey and cheese
in between. Cheese puffs can also be substituted for toast at break
Ifyou'rebeingtreatedwithonlylonger-acting bloodsugar-lowering
agents, the question of random or even preplanned snacking is best
answeredby experimentation usingblood sugar measurements.
Creatinga CustomizedMeal Plan 179
Meal and Medication Adjustments
Although your blood sugars wiU respond best if you adhere to our re
strictions on carbohydrate, you'U find that you have considerable lee
waywhen it comes to planning the amount of protein for each meal,
provided that you don't havegastroparesis or another digestive disor
der and inject insuUn. At the initial meal-planning sessionwith your
physician or other health careprovider, youmayestimatethat youwiU
require perhaps 6 ounces of protein to satisfyyour appetite at lunch.
When you actuaUy try eatingsuch a lunch, you mayconcludethat this
amount ofprotein is either too much or too Uttle for your satisfaction.
This can readUy be changed, provided that you first adviseyour health
care provider, so that dosageof any blood sugar-lowering medication
you take may be adjusted accordingly. Once a comfortable amount of
protein has been establishedfor a meal,it should not changefrom day
to daybut, likethe carbohydrate, be heldconstant The predictabiUty
of blood sugar levels under this regimen depends, in part, upon the
predictabiUty of your eatingpattern.
Carbohydrate or Protein Juggling
Manypatients ask me if they can jugglecarbohydrateor protein from
one meal to another, keeping the totals for the dayconstant Suchan
approach doesn't work, for reasons that should be obvious by now,
and can be downright dangerous if you're taking medications that
lower blood sugar. Many patients who visit me for the first time after
readingthis book havetotaUy ignoredthis veryimportant point and
havefound it impossibleto achieve stableblood sugars.
Calcium Concerns
Somepeoplewho foUow mydietaryguidelines consumeconsiderable
amounts of fiber. Slow-acting carbohydrate foods that are especiaUy
high in fiber includesalads, broccoU, cauUflower, bran, and soybean
products. Fiberbinds dietarycalcium in the gut, causing a reduction
of calciumabsorption and potential depletionof bone mineral, which
contains 99.5 percent of our calciumreserves. The phosphorus pres
ent in proteins also maybind calcium sUghtly. Since I discourage the
use of milk and certain milk products (except cheese, yogurt, and
cream), whichare good sources of dietary calcium, the potential for
bone mineral depletionmayindeedbe real. This is a special problem
180 Treatment
for women, who tend to lose bone mass at an increased rate after
menopause. I recommenda calcium supplement to anyone who fol
lows our diet and doesn't use cheese, yogurt, or cream, especiaUy
women. Sincesome women rapidlylose calciumfrom their bones af
ter menopause,it makes sense to buUd up calciumstoresearlierin Ufe,
and to offset high-fiber and high-protein diets with extra calcium.
Calciumsupplementation,bythe way, is most important for growing
teenagers who foUow low-calcium diets. Calcium supplements with
vitamin D, magnesium, and manganese are more effective, as these
aid calciumuptake bybone —and the magnesiumhelps counter the
potentiaUy constipatingeffect of calcium takenbyitself. Calciumsup
plementation also facUitates weight loss by sUghtly elevating your
metaboUc rate, but that doesn't mean that the more you take, the more
weight you'U lose. Nor does it meanthat if youarein the minority of
diabetics who are tryingto gainweightyou should avoidcalcium.
I recommend calcium citrate because it is weU absorbed in the gut
and inhibits the formation of kidney stones. One study of calcium
supplementation suggests the equivalent of at least 1,000 mgfor every
10ounces of protein consumed. Calcium supplements are best taken
with each meal. Calcium tablets taken at bedtime are often effective
in reducing the frequency of nocturnal muscle cramps in the legs.
Sedentary and thin people lose more bone calcium over a lifetime
than do physicaUy active people. Exercise buUds bone just as it buUds
The guidelines setforth inthis chapter should beadequate for you to
create your ownmeal plan, but I don't wantto leave youwithanyun
certainty as to how it is done. I have, therefore, Usted below 3 days'
worth of breakfasts, lunches, and suppersto give you an idea of howI
do it. These meals should serve as a starting point. You may want to
overhaul them entirelyto reflect your favorite foods. If, for example,
you prefer canned salmon to frankfurters, just substitute a smaU can
{SlA ounces) of salmon for the two 1.7-ouncefrankfurters in the lunch
of Day One.
The carbohydrate content of eachmeal reflects our 6-12-12 guide-
fines. If you're going to maintain normalbloodsugars, thenwhatever
amounts of carbohydrate youusemust remainrigidand reaUy should
Creatinga CustomizedMeal Plan 181
reflect maximum but constant amounts. (SmaU chUdren should theo-
reticaUy consume less, but this may pose problems of compliance.)
That said, however, exceeding or diminishing carbohydrate aUoca-
tionsby 1-2 gramsper mealfor adultswiU not makea great difference
in your blood sugars — remember the Laws of SmaU Numbers. Aside
fromtheseconstraints, youare otherwise limitedonlybyyour imagi
nation. The protein content of meals, on the other hand, is completely
up to you, provided that you don't take insuUn andhave a digestive
disorder. For the foUowing examples, I'vearbitrarUy assumed certain
amounts of protein that maybe too muchor too littleto satisfy your
desires; you wiU want to experiment to determine your own prefer
ences. Remember, however, that protein, likecarbohydrate, should be
kept constant fromone dayto the next for anygiven meal.
Let's assume that you've negotiated a mealplan,and the amounts of
assigned carbohydrate and amounts of protein that you think wiU sat
isfyyou are as foUows:
Breakfast: 6 grams carbohydrate, 3 ouncesprotein
Lunch: 12grams carbohydrate, 4 ounces protein
Supper: 12grams carbohydrate, 5 ounces protein
Note that none of the nine meals to foUow adds up precisely to
theseguideUnes for total carbohydrate and protein,yet aU of themare
quite closeand thus acceptable. Notealsothat I usuaUy don't list bev
erages. This is simply because most acceptablebeveragescontain nei
ther carbohydrate nor protein and maytherefore be ignored in our
computations. Remember, however, that every tablespoon of cream
for your coffee or tea contains0.4gramsof carbohydrate.
Day One
MushroomOmelet with Bacon(page398)
1 Bran-a-Crisp with butter
Green CabbageCole Slaw(page 406), 1serving
1G/G crispbread with mustard or butter
2 frankfurters
Carbohydrate Protein
(grams) (ounces)
3.1 2.8
i& LQ
7.1 3.8
3.0 0.2
11.8 3.6
%cup mbced saladwith oil,vinegar, and spices
2 tablespoons crumbledbluecheese on salad
Pan-FriedSwordfish with GingerScallion Butter
(page 422)
Day Two
4.0 -
0.4 0.7

12.0 5.5
Low-carbohydrate pancakes(page401)
2 sausage patties, 1ounce each
2 cups saladwithvinegar-and-oil dressing,
sprinkledwithgratedcheese 12.0 0
1 small can tuna (3V4 ounces) mixed with
1tablespooneachmayonnaise and choppedcelery 0.3 3.25
1 sliceAmerican cheese (place on top of tuna
and heat in microwave for tuna melt) 0.67 0.67
12-ounce bottle Blatz Cream Ale Assume 0 Assume 0
total 12.97 3.92
Quiche Lorraine (page437), %serving
ChocolateSouffle (page440)
Day Three
Carbohydrate Protein
Breakfast (grams) (ounces)
2 ounces smoked Nova Scotia salmon

2 G/G crispbreads 6.0 0.5
2 cheese puffs (page 178) U. LI
TOTAL 7.3 3.8
Creatinga Customized Meal Plan
Avocado Spread(page435), 1serving
Half a red or greenbell pepper, cut into strips
3Vi ounces hamburger meat
1tablespoonBacos brand soybaconbits
(kneadinto hamburgerbeforecooking)
1 medium artichoke, boiled, served with melted butter
4Vi ounces any meat, fish,or poultry, cookedas youlike -
6.6 0.4
3.8 -
- 3.5

12.4 4.4
12.4 0.5
~ 4,5
12.4 5.0
To any one of these meals you couldadd as a dessert a serving of
sugar-free JeU-0 brand gelatin (without maltodextrin), which would
not appreciably affect your carbohydrate aUocations. Again, you are
limited only by your imagination, and there are countless different
mealsyou cancreate that addup to no morethan6 or 12gramsof car
bohydrateand 3,4,5, or more ouncesof protein.
Weight Loss — If You're Overweight
Weight loss can significantly reduce your insulin resistance.
You may recaU from Chapter 1 that obesity, especiaUy ab
dominal (truncal, or visceral) obesity, causes insulin resis
tance and thereby can play a major role in the development of both
impaired glucose tolerance and type 2 diabetes. If you havetype 2 di
abetes and are overweight, it is important that weight loss become a
goal of yourtreatment plan. Weightreduction canalso slowdown the
process of beta ceU burnout by making your tissues more sensitive to
the insulin you stiU produce, aUowing you to require (and therefore to
produce or inject) lessinsuUn.
It may even be possible, under certaincircumstances, to completely
reverse your glucose intolerance. Long before I studied medicine, I
had a friend, Howie, who gained about 100pounds over the course of
a few years. He developed type 2 diabetes and had to take a large
amount of insulin (100 units daUy) to keep it under control. His
physician pointed out to him the likely connection betweenhis dia
betes and his obesity.To my amazement, during the foUowing year, he
was ableto lose 100pounds. At the end ofthe year, he had normal glu
cose tolerance, no need for insulin, and a newwardrobe. This kind of
success may only be possible if the diabetes is of short duration, but it
is certainlyworth keepingin mind — weight losscan sometimes work
Beforewe discuss weightloss, it makessenseto considerobesity,be
cause if you don't understand why and how you are overweight or
obese, it wiU be somewhat more difficult to reverse the condition.
Weight Loss —If You*re Overweight 185
When I see a veryoverweightperson, I don't think, "He ought to con
trol his eating."I think, "He has the thrifty genotype."
What is the thrifty genotype?
The hypothesis for the thriftygenotype was first proposed by the
anthropologist James V. Neel in 1962 to explain the high incidenceof
obesityand type 2 diabetesamong the Pima Indians of the southwest
ern United States. Evidence for a genetic determinant of obesity has
increased overthe years. Photographs of the Pimas froma centuryago
showa leanand wirypeople. Theydid not knowwhat obesitywasand
in facthad no word for it in their vocabulary.
Their foodsupplydiminished in theearlypart of the twentiethcen
tury, something that had occurred repeatedly throughout their his
tory. Now, however, they weren't faced with famine. The Bureau of
IndianAffairs providedthemwith flourand corn, and an astonishing
thing happened. Theseleanand wirypeopledeveloped an astronom
icalincidence of obesity— close to 100 percent of adult Pima Indians
today aregrossly obese, witha staggering incidence of diabetes. Today
half of adult Pimas in the United States are type 2 diabetics, and 95
percent of those are overweight. Since publicationof the first edition
of thisbook,manyPimachUdren have become obese, type2diabetics.
AsimUar scenario isnowplaying out across thecountryin the general
population. The pace may be slower,but the result is simUar.
Whathappened to thePimas?* Howdidsuch apparently hardyand
fit people become so grossly obese? Though their society was at least
in part agrarian, theylived in the desert, where drought was frequent
and harvests couldeasUy faU. Duringperiodsof famine, thoseof their
forebears whose bodies were not thrifty or capable of storingenough
energy to survive without food died out. Those who survived were
those whocouldsurvive longperiods without food. Howdid theydo
it?Although it may be simpUfying somewhat, the mechanism essen-
tiaUy works like this: Those who naturaUy craved carbohydrate and
consumed it whenever it was avaUable, even if they weren't hungry,
would have made more insuUn and thereby stored more fat. Add to
thisthe additional mechanism of thehigh insulin levels caused byin
herited insulin resistance, and serum insulin levels would have be-
*For more information on the Pimas, visit http://diabetes.niddk.nih.gov/dm/
186 Treatment
come great enough to induce fat storage sufficient to enable them to
Uve through famines. (See Figure 1-1.) Truly survival of the fittest —
providedfamines wouldcontinue.
A strain of chronicaUy obese micecreated inthe early 1950s demon
strates quitevividlyhowvaluable thrifty genes canbein famine. When
these mice areaUowed an unlimited food supply,they baUoon and add
asmuch ashalf again the body weight of normal mice.Yet deprivedof
food, these mice can survive 40 days, versus 7-10 days for normal
More recent research on these chronicaUy obese mice provides
some tantalizingly direct evidence of the effectathrifty genotype can
have upon physiology. Innormal mice, ahormonecaUed leptinis pro
duced in the fat ceUs (also a hormone human fat cells produce, with
apparently simUar effect). The hormone tends to inhibit overeating,
speed metabolism, andact as amodulator of bodyfat. Agenetic"flaw"
causesthe obese mice to make a less effective form of leptin. Experi
ments showed that when injected with the real thing they almost in
stantly slimmed down. Not onlydidtheyeat less but theylostas much
as 40percent of their bodyweight, their metabolismsped up,andthey
became much more active. Many were diabetic, but their loss of
weight (andthe change in the ratio of fat to lean body mass) reversed
or even "cured" their diabetes. Normal mice injected with leptin also
ateless,became more active, andlost weight, though not as much. Re
search on humans hasnot advanced sufficientlyto provide conclusive
evidence that the mechanism is the same in obese humans, but re
searchers beUeve it is at least equivalent and probably related to more
thanone gene, andto different gene clusters in different populations.
In a fuU-blown famine, the Pima Indian's abUity to survive long
enough to find food isnothing shortof ablessing. Butwhensatisfying
carbohydrate craving issuddenly justamatter of going to the grocery
or malting bread, what was once an asset becomes avery serious lia
Although current statistics estimate slightly more than60 percent
of the overaU population of the United States as chronicaUy over
weight, there isevengreater reason tobeconcerned, because thenum
ber has been increasing each year. Some researchers attribute rising
obesity in the United States at least in part to increasing numbers of
former smokers. Others attribute it to the recent increasein carbohy
drate consumptionby thosetryingto avoid dietary fat. Whatever the
reasons, overweight andobesitycanlead to diabetes.
Weight Loss—IfYou're Overweight 187
The thrifty genotype has its most dramatic appearance in isolated
populations like the Pimas, which have recently been exposed to an
unlimited foodsupplyafter miUennia of intermittent famine. The Fiji
Islanders, for example, wereanotherlean,wiry people,accustomedto
the rigors of paddUng out against the Pacific to fish. Theirdiet,highin
protein andlowin carbohydrate, suitedthem perfectly. Afterthe onset
of the tourist economythat foUowed World War II, their diet changed
to our high-carbohydrate westerndiet, and they too began (and con
tinue) to suffer fromahigh incidence of obesityand type 2 diabetes.
The same is true of the AustraUan Aboriginesafterthe Aboriginal Ser
vice beganto provide them with grain. Ditto for South Africanblacks
who migrated from the bush into the big cities. Interestingly, a study
that paidobese,diabetic South African blacks to goback to the coun
trysideand return to their traditional high-protein, low-carbohydrate
diet found that they experienced dramatic weight loss and regression
of their diabetes.
It's clear that thrifty genotypes work in isolated populations to
make metabolismsupremelyenergy-efficient, but what happens when
the populations have unrestrictedaccess to high-carbohydrate foods?
It would appear that the mechanism of the thrifty genotype
works something like this: Certain areas of the brain associated with
satiety— that sensation of being physicaUy and emotionaUysatisfied
by the last meal — may have lower levels of certain brain chemicals
known asneurotransmitters. A number of years ago, Drs. Richardand
Judith Wurtman at the Massachusetts Institute of Technology (MIT)
discovered that the level of the neurotransmitter serotonin is raised in
certain parts of the hypothalamus of the animal brain when the ani
mal eats carbohydrate, especiaUy fast-acting concentrated carbohy
drate like bread. Serotonin is a neurotransmitter that seems to reduce
anxiety as it produces satiety. Other neurotransmitters such as dopa
mine, norepinephrine, and endorphins can also affect our feelings of
satiety and anxiety. There are now more than one hundred known
neurotransmitters, and many more of them may affect mood in re
sponse to food in ways that are just beginning to be researched and
In persons with the thrifty genotype, deficiencies of these neuro
transmitters (or diminished sensitivity to them in the brain) causes
both a feeling of hunger and amUddysphoria— often a sensation of
anxiety, the opposite of euphoria. Eating carbohydrates temporarily
causesthe individual to feelnot only lesshungry but alsomore at ease.
188 Treatment
A frequent television sitcom scenariois the woman just dumped by
her boyfriend who plopsdown on the couchwith apieor half agaUon
of ice cream, a spoon, and the intention of eating the whole thing.
She's not reaUy hungry.She's depressed andtryingto make herself feel
better.She's indulgingherself, we think, rewarding herselfin away for
enduring one of life's traumas, and we laughbecause we understand
the feeling. But there is a very real biochemical mechanism at work
here. She craves the sugar in the pieor the icecreamnot because she's
hungry but because she knows, consciously or not, that it reaUy wiU
make her feel better. Contrary to popular belief, the fat in the ice
cream or in the crust of the pie doesn't make much of a difference. It's
the carbohydrate that wiU increase the level of certainneurotransmit
ters in her brain and make her feel better temporarily. The side effect
of the carbohydrate is that it also causes her blood sugar to rise and
her body to make more insulin; and, as she sits on the couch, the ele
vation in her seruminsulinlevel wUl facUitate the storage of fat.
On television the actress may never get fat. But for the real-life
woman, high serum insulin levels from eating high-carbohydrate
foodswiU cause her to crave carbohydrate again. If sheis atype 1dia
beticmakingno insulin, she'U have to inject alot of insulinto get her
blood sugar down, with the sameeffect — more carbohydrate craving
and buUdingup of fat reserves.
There may be many mechanisms by which the thrifty genotype can
cause obesity. The most common overt cause of obesityis overeating
carbohydrate, usuaUy over aperiod of years. Unfortunately, thiscanbe
avery difficult type of obesity to treat.
If you're overweight, you're probably unhappy with your appear
ance, and no less with your high blood sugars. Perhaps in the past
you've tried to foUow a restricted diet, without success. GeneraUy,
overeating foUows two patterns, and frequently they overlap. First is
overeating at meals. Second is normal eating at mealtime but with
episodic"grazing." Grazing canbe anything from nibblingand snack
ingbetweenmealsto eating everything that doesnot walkaway. Many
of the people who foUow our low-carbohydrate diet find that their
carbohydratecravingceases almost immediately, possiblybecauseof a
reduction in their serum insulin levels. The addition of strenuous ex-
Weight Loss—IfYou're Overweight 189
ercise sometimes enhances this effect. Unfortunately, these interven
tions don't work for everyone.
If you're a compulsive overeater, if you just can't stopyourself from
eating, and are addicted to carbohydrate, you may not be able to ad
hereto our diet without somesort of medical intervention (see Chap
ter 13). Carbohydrate addictionisjust asrealasdrugaddiction,and in
the case of the diabetic, it can likewise have disastrous results. (In ac
tual fact, excess body weight kiUs more Americans annuaUy from its
related complications than aU drugsof abuse combined, including al
Youneed not despair of never losing weight, however. I have seen a
number of "diet-proof" patients over the years get their weight down
and blood sugars under control. Over the last several years, medical
science has gained a much more sophisticated understanding of the
interactions of brain chemicals (neurotransmitters) that contribute to
emotional states such as hunger and mood. Many relatively benign
medications have been successfuUy applied to the temporary treat
ment of compulsive overeating. There is no doubt that when used
properly, many appetite suppressants are quite effective in helping
peopleto loseweight.If you simplycannot loseweight,it maybe help
ful to discusswith your physicianmedicinesthat maybe of use to you.
I haveused more than 100different medications with my patients and
have found many of them to be of great value for treating carbohy
drate addiction.
There is, however, a catch to this method. Over the years, I have
found that none of these medications workscontinuaUyfor more than
a few weeks to a few months at a time, a fact that many if not most
medical and diet professionals may be unaware of.
I developed a reasonablysuccessful method for prolonging effec
tiveness ofsome by rotating them weekly, so that from one weekto the
next a different neurotransmitter would be caUed into action to pro
vide the sensation of satiety. I found that about eight different med
ications, changed everyweek for eight weeks, and then repeating the
cycle, would perpetuate the effect for as long as people continued to
take them. At one point this looked to be a very promising means to
help get weight off and keep it off. I even acquired a patent for the
technique. Over time, however, I found severalsignificant reasons not
to continue pursuing this route. The most insurmountable of these
190 Treatment
was that it was just toodifficult for mostpeople to foUow their normal
regimen of diabetes medications whUe at the same time changing
theirregimen ofappetite suppressants from week toweek. Add tothat
thedifficulty of working with apatient over a number of weeks just to
findeight medications that worked for themandcould be rotated.
What I did discover during aU this trial and error wereseveral effec
tivemethodsof curbingovereating. The results mypatientshave had
withthemaresosignificant that I've devoted thewhole next chapterto
Reducing Serum Insulin Levels
Another group of type 2 diabetics has a common story:"I was never
fat until after my doctor started me on insulin." UsuaUy these people
havebeen foUowing high-carbohydrate diets and so must inject large
doses of insuUnto effecta modicum of blood sugar control.
InsuUn, remember, is the principal fat-buUding hormone of the
body. Although a type2diabetic maybe resistant to insuUn-facUitated
glucose transport (frombloodto ceUs), that resistance doesn't dimin
ish insulin's capacity for fat-buUding. In other words, insulin can be
great at makingyou fat eventhough it maybe, for those with insulin
resistance, inefficient at lowering your blood sugar. Since excess in
sulin is a cause of insuUnresistance,the more you take, the more you'U
need, and the fatter you'U get. This is not an argument against the use
of insulin; rather it supports our conclusionthat high levels of dietary
carbohydrate — which, in turn, require large amounts of insulin —
usuaUy makeblood sugar control (and weight reduction) impossible.
I have witnessed, over and over, dramatic weight loss and blood
sugar improvement in peoplewhohavemerelybeen shown howto re
duce their carbohydrate intake and therefore their insulin doses. Al
though this is contrary to common teaching, you need only visit the
reader reviews of earUer editions of this book to read the simUarexpe
riences of many readers.*
Several oral insulin-sensitizing agents, whichwewUl discussin de-
taUin Chapter 15,can alsobe valuabletools for facUitating weight loss.
They work by making the body's tissues more sensitive to the blood
sugar-lowering effectof injectedor self-madeinsulin. Asit then takes
* At www.amazon.com, www.amazon.co.uk, and www.diabetes-bookcom.
Weight Loss—IfYou're Overweight 191
less insulin to accomplish our goal of blood sugar normalization,
you'U have less of this fat-building hormone circulating in your body.
I have patients using these medications whoarenot diabetic, and they
work in a simUar way: the body is more sensitive to insulin, so it needs
to produceless, and there is, again, less of it present to build fat. One
mayalso have less of a sense of hunger, andless lossof self-control.
Increasing Muscle Mass
The above suggests what wehave beenadvocating aU along—a low-
carbohydrate diet. But what do you do if this plus one of the above
medications does not result in significant weight loss? Another step
is muscle-buUding exercise (Chapter 14). This is of value in weight
reduction for several reasons. Increasing lean body weight (muscle
mass) upgrades insulin sensitivity, enhancing glucose transport and
reducing insulin requirements for blood sugar normalization. Lower
insulinlevels facilitate loss of storedfat. Chemicals produced during
exercise (endorphins) tend to reduce appetite, as do lower serum in
sulin levels. Peoplewho haveseen results from exercise tend to invest
more effort in lookingeven better (e.g., by not overeating, and per
haps exercising more). Theyknowit can be done.
Now suppose you have been foUowing our low-carbohydrate diet,
have been conscientiously "pumping iron,"and are, in effect, "doing
everythingright."What else can you do if you have not lost weight?
WeU, everyone has some level of caloricintake belowwhich they wiU
lose weight. Unfortunately, the "standard" formulas and tables com
monly used by nutritionists set forth caloricguidelines for theoretical
individuals ofa certain age, height, and sex,but not for real people like
us. The only way to find out how much food you need in order to
maintain, gain, or loseweight isbyexperiment. Here is an experimen
tal plan that your physician may find useful. This method usuaUy
works, and without counting calories.
Beginbysetting an initial target weightand a reasonabletime frame
in which to achieveit. Usingstandard tables of "ideal body weight" is
of littlevalue, simplybecause they give a verywide target range. This
192 Treatment
is because some people have more muscle and bone mass for a given
heightthanothers. The highendof the ideal weight for agiven height
on the MetropoUtan Life Insurance Company's table is 30 percent
greater thanthe lowend for the same height.
Instead, estimateyour target weight by looking at your body in the
mirror after weighing yourself. (It pays to do this in the presence of
your healthcare provider, because he/she probably has more experi
ence in estimatingthe weight of your body fat.) If you can grab hand-
fuls of fat at the underside of your upper arms, around your thighs,
around your waist, or overyour beUy, it is pretty clear that your body
is set for the next famine. Your estimate at this point need not be ter
ribly precise, because as you lose weight your target weight can be
reestimated. Say, for example, that youweigh200 pounds.You andyour
physician may agree that areasonable target would be 150 pounds. By
the time you reach 160pounds, however, you may havelost your visi
ble excess fat — sosettle for 160 pounds. Alternatively, if you still have
fat around your beUy when you get down to 150 pounds, it won't hurt
to shoot for 145 or 140 as your next target, before making another
visual evaluation. GraduaUy you home in on your eventual target, us
ing smaller and smaUer steps.
Once your initial target weight has been agreedupon, a time frame
for losing the weight should be established. Again, this need not be ut
terly precise. It's important, however, not to "crash diet." This may
cause a yo-yo effect by slowingyour metabolism and making it diffi
cult to keep off the lost bulk. Bear in mind that if you starveyourself
and lose 10 pounds without adequate dietary protein and an accom
panyingexercise regimen, you may lose5 pounds of fat and 5 pounds
of muscle. If you gain back that 10pounds from eatingcarbohydrate
and stiU are not exercising, it may be all fat. After crash dieting, once
you've reachedyour target, you may go right back to overeating. I like
to have my patients foUow a gradual weight-reduction diet that
matches ascloselyas possible what they'U probably be eatingafter the
target hasbeen reached. In other words, once your weight has leveled
off at your target, you stayon the samediet you foUowed whUe losing
weight — provided, of course, that you don't continue losingweight.
This way you've gotten into the habit of eatinga certainamount, and
you stick to this amount, more or less, for life.
To achieve this, weight loss must be gradual. If you aretargeted to
lose 25 pounds or less, I suggest a reduction of 1 pound per week. If
Weight Loss — If You're Overweight 193
you're heavier, you maytry for 2 pounds perweek. If just cuttingthe
carbohydrate results in a more rapid weight loss, don't worry— just
enjoyyour luck. This hashappenedto anumber of my patients.
Weighyourselfonceweekly— stripped,if possible, on the samescale,
andbeforebreakfast. Pickaconvenient day, andweighyourself on the
same day eachweek at the same time of day. It's counterproductive
andnot very informativeto weighyourselfmore often. SmaU, normal
variations in body weight occur fromdayto day and can be frustrat
ing if you misinterpret them.* GeneraUy speaking, you won't lose or
gain apound of body fat in aday. Continue on your low-carbohydrate
diet, with enough protein foods to keep you comfortable.
Let's say that your goal is to lose 1pound everyweek.Weighyour
self afterone week. If you'velost the weight, don't change anything. If
youhaven't lost the pound, reduce the protein at anyone mealby one-
third. For example, if you've been eating 6 ounces of fish or meat at
dinner, cut it to 4 ounces. Youcan pickwhichmeal to cut. Checkyour
weight one week later. If you have lost a pound, don't change any
thing. If you haven't, cut the protein at another meal by one-third. If
you haven't lost the pound in the subsequent week,cut the proteinby
one-third in the one remaining meal. Keep doing this, week by week,
until you are losingat the target rate. Never addback any proteinthat
you have cut out, even if you subsequently lose 2 or 3 pounds in a
If you'vemanaged to lose at least 1 pound weekly for many weeks
but then your weight levels off, this is a goodtime for your physician
to prescribe the special insulin resistance-lowering agents described
in Chapter 15. Alternatively you can just start cutting protein again.
Continue this until you reach your initial target or until your visual
evaluation of excess body fat teUs you that further weight loss isn't
necessary. The average nonpregnant, sedentary adult with an ideal
body weight of 150 pounds requires about 11.5 ounces of high-
quaUty protein food (i.e., 69 grams of pure protein) daUy to prevent
proteinmalnutrition. It is therefore unwiseto cut your protein intake
*This is especially true for many menstruatingwomen, who retain more water
during the weekbeforetheir periods.
tThis may not work for girls or women with polycystic ovarian syndrome
(PCOS). TheymayfaU to loseweightevenon a near-starvation diet (seeAppen
dix E).
194 Treatment
much below this level (adjusted for your own ideal body weight). If
you exercise strenuouslyand regularly, youmayneed much more than
this in order to buUdyour muscles. GrowingchUdren also need more
protein. Once you'vereachedyour target weight, do not add back any
food. You wiU probably have to stay on approximately this diet for
manyyears, but you'U easUy becomeaccustomed to it. If you required
one of the appetite-reducing approaches described in the next chap
ter, do not discontinue it.
Reduce Diabetes Medications While Cutting Protein
or Losing Weight
WhUe you'relosingweight, keepchecking blood sugarsat least 4times
daUy, at least 2 daysa week. If theyconsistently drop belowyour target
value for even a fewdays, adviseyour physician immediately. It wiU
probablybe necessary to reducethe dosesof anyblood sugar-lowering
medications you may be taking. Keeping track of your blood sugar
levels as you eat less and loseweight is essential for the prevention of
excessively lowblood sugars.
Increased Thrombotic Activity During Weight Loss
During weight loss, many people unknowinglyexperience increased
clumping of the smaU particlesin the blood (platelets) that form clots
(thrombi). This can increase the risk of heart attack or stroke. Your
physician may thereforewant you to take an 80 mg chewable aspirin
once daUy during a meal to reducethis tendency. The aspirin should
be chewedmidwaythrough a meal to reduce the possibUity of irrita
tion to the stomach or intestines. Rinseyour mouth with water or diet
soda after chewing aspirin to prevent inflammation ofyour gums. Al
ternatively, instead of aspirin you can use vitamin E in the form of
gamma tocopherol or mixed tocopherols. The dosing would be 400
mg one to three times daUy dependingupon your size. It need not be
taken during meals, as it won't irritate your gastrointestinal tract.
Elevated Serum Triglycerides During Weight Loss
When you're losingweight, fat is"mobUized" for oxidation— i.e., to
be burned — and it wiU appear in the bloodstream as triglycerides. If
you see elevatedserum triglyceride levels as you're losing weight, it's
Weight Loss — If You're Overweight 195
not something to worry about. Your triglyceride levels wiU drop as
soon as weight loss levelsoff.
Supplemental Calcium May Help
There is evidence that dietary calcium and to a lesser degree calcium
supplements (1,000-3,000 mgdaUy) mayfacUitate weight loss by in
hibiting the accompanying slowdown in metaboUsm that mayoccur
whenyouloseweight. If usedfor thispurpose, the supplementshould
not contain vitamin D, as it wiU counteract the effect on weight loss.
How to Curb Carbohydrate
Craving or Overeating
Thediet plandescribed in this book should make it possible for
virtuaUy any diabetic to achieve normal blood sugars. There
are, however, three exceptions. The first, as I've mentioned, is
the presence of gastroparesis, or the partial paralysis of the stomach,
and other ailments that canimpair stomach-emptying. These may in
clude hiatal hernia, stomach or duodenal ulcers, a "tonic" (tight)
stomach, gastritis, duodenitis, and scleroderma, among others. The
next is infection. The third is the inabUity to control food intake, but
especiaUy carbohydrate intake. Because of the thrifty genotype, we
should expect to find this condition in many type 2 diabetics. Indeed,
about 25 percent of my type 2 patientsfindit extremelydifficult to re
main on a low-carbohydrate diet — or indeed any kind of structured
diet. Typical scenarios include snackingwhen bored, eating bread in
restaurants for no better reason than that it's on the table, and eating
everythingon your plate regardless of anyactual hunger if you happen
to be givenatoo-large portion, often at restaurants. Others may eat a
whole pint, and some even aquart, of icecreameverynight, often be
causethey feel they have nothing elseto do. At least 10percent of my
type 1diabetic patients have suchproblems, but their problems, when
they occur, have a more devastating effect on blood sugars. These are
the people who rapidly develop retinopathy, numb feet, kidney dys
function, and so on.
Several years ago, I recaUed how, in medical school, I had been
taught a technique for self-hypnosis in order to avoid faUing asleep
when we had boring speakers and the lights were turned off. I won
dered if perhaps the same technique might be helpful to those patients
who just couldn't seem to stick to a low-carbohydrate diet or any
HowtoCurb Carbohydrate Craving orOvereating 197
other. I decided to get in touch with the physician, Herbert Spiegel,
MD, who had taught me autohypnosis.
Remarkably, he was stUl in the same office at the same telephone
number asmore than twenty years before. Tomy amazement, he told
me that he routinely used this technique to treat people with eating
problems. Put this under the category of reasonably important dis
coveries I had missed. In fact, with his son, David, he had even written
abook, Trance andTreatment: Clinical Uses ofHypnosis. It includes a
chapter on just this subject. Since then, I've referred manypatients to
him to learn this technique, andwe've hadsuccess in helpingpeople
break the cycle of carbohydrate addiction. (Dr. Spiegel uses this
method for altering a number of different behaviors, not just weight
loss or the propensity of elderly medical students to snore duringsop
orific lectures.)
The whole cycle of going into a hypnotic state, giving yourself a
message, and coming out of the state takes initiaUy about 1 minute.
Once you have some experience, it onlytakes about 20 seconds to com
plete. The technique only works if you're hypnotizable. A quaUfied
medical hypnotist quite readUy candetermine whether you areasuit
able subject. One of the things the specialist looks for ishowhighyou
can roU up your eyes as you attempt to look toward the top of your
head.AdditionaUy, the techniquewUl only work if you perform auto
hypnosis at least10times daily (for adaUy total of about 3Vi minutes).
The sectionbelow, which is adaptedwith permission from Dr.Spiegel's
book, is the handout that our eminent consultant, Dr.Spiegel, gives to
his patients with eating problems. This handout is a reminder of what
they've been taught in his office.
Sit or lie down. At first, being in a quiet place can help. To
yourself, count to three.At one you are goingto do one thing, at
two, you wiU do two things, at three, you wiU do three things.
* Reproduced (with minor modifications by me) from Trance and Treatment:
Clinical Uses ofHypnosis, byHerbert Spiegel, MD,and DavidSpiegel, MD,Amer
ican Psychiatric Press, 1987.
198 Treatment
1. Without moving your head, look upward toward your
eyebrows, attthe way up.
2. Close youreyeUds andtake adeep breath.
3. Exhale; let your eyesrelax andlet your body float.
As you feel yourselffloating, you permitonehandorthe other to
feel like a buoyant baUoon and aUow it to float upward. As it
does, your elbowbends and your forearm floats into a vertical
position. Sometimes you may get a feeling of magnetic puU on
yourhandasit goes up.When yourhandreaches thisvertical po
sition, it becomes a signal for you to enter a stateof meditation
and increase your receptivity.
In this state of meditation, you concentrate on this feeling of
imaginary floating and at the same time concentrate on these
three messages:
1. Formy body, overeating is a poison.
2. I need my body to live.
3. I owe my body this respectand protection.
In the beginning, do these exercises as often as 10 different
times a day,* preferably every 1-2 hours. At first, the exercise
should take about a minute, but it will come more rapidly with
As you meditate, reflect on the impUcations of these critical
points and then bring yourself out of this state of concentration
by counting backwards in this manner: Three, get ready; two,
with your eyeUds closed, roU up your eyes (do it now); and one,
let your eyeUds open slowly. Then, when your eyesare back into
focus, slowly make a fist with the hand that is up and, as you
open the fist slowly, your usual sensationand control returns. Let
your hand float downward. This is the end of the exercise, but
you retaina general, overaU feeling of floating.
By doing this exercise (at least) 10 different times each day,
you can float into this stateof buoyant repose. Giveyourself an
island of time. Twenty seconds, 10 times a day, in which to use
* I have some patients who must do this 15 times daUy, including once before
each meal.
t Requiringonlyabout 20seconds eachtime.
HowtoCurb Carbohydrate CravingorOvereating 199
this state of extra receptivity to reimprint these three points. Re
flect upon it, then float backto yourusual state of awareness and
get on with what you ordinarUy do.
Now, suppose an hour or two passes and you want to do the ex
ercise. You don't havethe privacy and you don't want to attract
attention. Here's the wayto camouflage. There are two changes.
First, you close your eyes and then roU your eyes up, so that the
eye roU is private. Second, instead of your hand coming up as
done in the hypnosis session [with the hypnotist], let it come up
and touch your forehead. To an outsider, the exercise looks as
though you are in deep thought. In 20 seconds you can shift
gears, establish this extra receptivity, reimprint the critical
points, and shift back out again.
You might be sittingat a desk, a table, or may be in a confer
ence,in which case you leanoveron your elbowwith your hand
already on your forehead, you closeyour eyes,roU them up, and
shift into the brief meditative state.
By doing this basic or camouflaged exercise every day, every
one or two hours, you estabUsh aprivate signal systemsothat you
are ever alert to the messages you are sending yourself and the
commitment you are makingto yourselfand your good health.
It is possiblebut unlikely that you wiUbe ableto master this technique
without training from a professional medical hypnotist. Sufficient
training should be possibleto accomplishin asingleoffice visit with a
doctortrained in medical hypnotism.I stress the necessity of usingthe
services of a doctor because I have had patients who, upon visiting
nonphysician hypnotherapists, were convinced (you might even say
conned) that many officevisits werenecessary, and spent considerable
sums but faded to learnthe technique. (If the word"hypnotism" auto-
maticaUy conjures up images of charlatans and carnival sideshows in
your mind, this may be areason.)
You should be able to locate a competent medical hypnotist by
phoning the department of psychiatry at the nearest medical school or
200 Treatment
teaching hospital. Ask for the secretary/assistant to the chairman of
the department, and then ask who their top MD hypnotherapist is.
Your insurance may or may not cover the visit, depending on your
plan, but it almost certainly wUl not pay for a nonphysician hyp
When youvisit the MDhypnotherapist, bringeither this book or a
photocopy of the above paragraphs sothat he/she wUl knowexactly
what youare seeking. If youUve withinareasonable distance of New
York City, Dr. Spiegel's office is located on East Eighty-eighth Street
and his office phone number canbe located in the Manhattan direc
tory. He is the hypnotherapist to whom I refer my patients.
In Trance and Treatment he is very emphatic on the foUowing:
Accept responsibUity for Your Eating Behavior. It is very tempt
ingto blameyoureating behavior on your parents, yourwife, the
mayor, Watergate, the moon, the tides. As soon asyou seethe ab
surdity of that you wiU realize that of aU the things you do in Ufe,
there is nothing in which you are more clearly 100 percent re
sponsiblethan your eatingbehavior. Reflecton the fact that most
of the things you do in life haveto take into account other con
siderations or other people, but in your eating behavior you are
in business for yourself.
I should note that Dr. Spiegel's three points, or messages, were de
veloped for the treatment of obeseovereaters, not necessarUy diabet
ics but certainly people at higher risk for developing type 2 diabetes.
For many of my patients, his three points hit the mark. For others, the
points they want to stress are more closely attuned to their personal
situations, and you cancustomize your approach asweU. I haveone
patient who is simultaneouslylosingvision and kidney function. This
patient's personal points are to curb overeating in order to preserve
eyesight and to stayoff dialysis. Another teUs herself that she doesn't
want to be like the ladywho Uved across the street fromher when she
was a kid — the woman was diabetic and had both her arms and legs
amputated. Another patient is a fund-raiser — his points areattuned
to the greater likelihoodof people makingdonationsto someone who
is slim and trim rather than obese. Makes sense.
Some peoplehaveadifficult time remembering to hypnotize them
selves every 1-2 hours, so for these people, I recommend an alarm
HowtoCurb Carbohydrate CravingorOvereating 201
watch (which youcanpurchase quite reasonably) that canbesetto go
offevery hour. Getonethat has avibration mode ifyoudon't wantto
be beeping regularly,* and you can use the camouflage technique if
your situation requires it.
For those who learn howto hypnotize themselves and do it 10-15
times a day, I'vefound that the success rate for curbingcarbohydrate
craving is about 80 percent. For those who hypnotize fewer than 10
times a day, the success rateisessentiaUy zero. I cannotoveremphasize
thevalue of engaging inautohypnosis when yousit downat a table for
a meal, especiaUy if you're in a restaurant and have not yet ordered.
I have patients who are walking around with normal blood sugars
only because they have been successful using this technique. It has
the added benefit of having no toxicity whatsoever, and it might be
used to change other behaviors (smoking, biting fingernaUs, and
so on).
Thereis, however, a major problem withautohypnosis. I find that
many patients either refuse to try it or eventuaUy stop doing it, even
when it works. The foUowing section offers asolution toovereating that
people have beenmorethanwiUing to pursue andcontinue to use.
I have a simple, patented technique that has hada very high success
rate for those of mypatients who otherwise have great difficulty con
trolling carbohydrate intake. I have been prescribing it for several
years but cannotyet be certain that it wUl work indefinitely. It relates
to the"runner's high" that people often experience during and after
exercise, but it doesn't require running.
You may already know that very strenuous, prolonged physical
exercise and climactic sexual activity cause the brain to produce en
dorphins, also known as the body's own opiates because they are pro
duced internaUy (or endogenously —so they're known in medical
*An excellent source for such products is e-PUl, LLC, 70 Walnut Street, Welles-
ley, MA, (800) 549-0095, orwww.medicalwatches.com. I recommend their Pager
Vibrating Multi-Alarm ($75.95). It will give asmany alarms asyoudesire if used
in the "repeat countdown" mode.
202 Treatment
circles as "endogenous opiates") and bind to receptors in the brain
that bind actual opiates, suchas morphine and codeine. Endorphins
cause a pleasant, relaxed feeling, simUar to that of narcotics, but to a
mUder degree andwithout producing atolerance.* Youmayhaveno
ticed that serious runners and many professional athletestend to pre
fer protein foods, don't crave carbohydrate, and don't become fat as
longas theycontinue theirsport. It wouldseemthat theirendorphins
prevent overeating andcarbohydrate craving without losing their ef
fect over time asdo traditional appetitesuppressants.
This technique involves amedication caUed naltrexone, whichwas
originaUy introduced as a treatment for narcotics addicts because of
its abUity, in large doses, to prevent addicts from getting highon nar
cotics. In large doses, naltrexone wiU block the brain's receptor sites
for endorphins, rendering them ineffective. However, when taken in
smaU doses, it also appears to raise endorphinlevels in the brain.
A verysmaU dose of naltrexone taken at bedtime appears to block
endorphin receptors for about 8hours. Thebrain maycompensate for
this by making more endorphins than usual that then keep working
throughout the foUowing day, whenreceptors are no longer blocked.
I've found that asmaUdoseof naltrexoneusedin this wayis very ef
fective in controUingcarbohydrate addiction and overeating in gen
eral forabout 73 percent of thosewho have triedit. Furthermore, just
likethe endorphinsmadeby athletes, naltrexone, thus far in my expe
rience, seemsto work for a prolonged period— perhaps indefinitely.
I'vehadonly five patients who discontinued naltrexone because of
uncomfortable side effects. These effects included tiredness, headache,
and difficulty concentrating on complex tasks. Such problems always
occur afterthe first doseor adosage increase. When they occur, I must
either lower the dose or discontinue the medication.
Onepatient whosnacked between dinner and bedtime also hadin
somnia. Since naltrexone made him tired, we were able to use it to
treat his insomnia andhis snacking simultaneously.
Naltrexone is suppUed in 50 mg tablets. For the low dose that I
prescribe, I use a compounding chemist to put naltrexone powder
intosmaU capsules — usuaUy about 4.5 mg.f
* In medical terms, "tolerance" refers to declining efficacy over time, so that
higher andhigher doses are required to produce thesame results.
fThe compounding chemists that most of my patients use are Rosedale
Pharmacy, phone (888) 796-3348, and Rockwell Compounding Associates,
Howto CurbCarbohydrate Craving or Overeating 203
Although in the pastI've prescribed naltrexone atvariousdosesand
at different times of the day, I changedthis to 4.5 mg at bedtime at the
suggestion of Dr. Bernard Bihari, who has prescribed it for ailments
other than overeating. This dosing method indeed appears to be the
most effective.
Side effects that some patients have caUed to my attention include
occasional headaches or a feeling of mental confusion that impairs
concentration on difficult tasks.These effectsareunacceptable, and so
I may start my patients on avery low dose — 0.5 mg capsules— and
increase the amount of each dose until we reach an effective level. One
must keep in mind, however, that smaUer doses are often more effec
tive than larger ones.
As with the other suggestions in this book, ask your physician to
give naltrexone a try. He should payparticular attention to the pack
age insert warningsagainst overdosing. For other remarkable uses for
low-dose naltrexone, visit the Web site www.lowdosenaltrexone.org.
I have patented this mode of using naltrexone,in order to encour
age its distribution by pharmaceutical companies in low doses. Such
companies arenot interestedin sellingproducts that arenot protected
by patents.
Another product has curbed overeating in about 65 percent of the
patients for whom I've recommended it. This is hoodia, a nonpre
scription extract of a South Africancactus. The brand I prefer is caUed
HoodiaXtra, 1,000 mg. It is avaUable online at www.hoodiaxtra.com
and www.diabet.es911.net. Other brands, usuaUyin smaUerdoses, may
be purchasedat health food stores, Rosedale Pharmacy, and at other
sites online. Since hoodia begins working within 30-60 minutes, it
should be taken about 1hour before an anticipated need. Thus, if you
usuaUy overeat at dinner and then snack thereafter, you might take
1-3 capsules before dinner and another 1-3 after dinner. Some
sources of hoodia supply a timed-release version. It is most effective
when taken on arising and again6-8 hours later.
I have seen no adverse side effects from this product, but I have
never used it in chUdren.
(914) 925-2304. (Compounding chemists are pharmacists with special training
in the precise mixing of pharmaceuticals and over-the-counter medications.
Theycan prepare capsules, powders, liquids, and evenointments, just as all phar
macists did when I was a child. There are many compounding chemists in the
United States and elsewhere. Most require a physician's prescription.)
204 Treatment
Because demand for hoodia now exceeds the supply, many phony
formulations are being marketed, especiaUy on the Internet. Tosecure
a reliableproduct, you would be wiseto use a known brand supplier
and a product without additivesthat may dUute its strength.
I believe that carbohydrate craving is truly an addiction. This ad
diction can be reinforcedby consumption of foods on our No-No list
(pages 152-153). So once you have discontinued them, I recommend
never trying them again or you wiU likelyrelapse. Even a smaU taste
can cause you to faU off the wagon— just as for smokers, alcoholics,
and other drug addicts.
You may recaU from our discussion of the Chinese restaurant effect
(page 97) that intact pancreatic beta ceUs make a hormone caUed
amylin. Amylin productionisbrought about in response to a meal,by
gut hormones caUed incretins. Since diabetics do not have beta ceUs
that function adequately, they make Uttle or no amylin and therefore
may not experience the degree of satietythat nondiabetics do. They
are therefore more likelyto remain hungry after a meal and thus to
overeat at meals and snack between meals.
Perhaps the most excitingclassof drugs to hit the market in many
yearsactuaUy solves this problem. Productsin this categoryare caUed
incretin mimetics (IMs) — that is, they imitate the effects of incretins
or of amylin. What is so special about them is that in my experience
they reaUy work, and they do so for about 90 percent of users, a very
high degree of success. At present these products are sold to lower
blood sugar after meals. Our appUcation to overeating is therefore
considered"off label" but is permitted by the FDA if prescribedby a
The most effective IMs,at this writing, must be administeredby in
jection one or more times daUy. The good news is that the needles are
tiny, so these injections are painless if you foUow our injection tech
nique (Chapter 16). A once-weekly version wiU be avaUable shortly
and an oral version (Januvia) is nowon the market. But start nowwith
the injections because the benefits are so great.
Howto CurbCarbohydrateCravingor Overeating 205
For example, I sawa newpatient recently whoweighed 286pounds
(130 kUograms) andhadanHgbAlc of 6.9 percent when we first met.
I started her on an IM, and over a period of sUghtly less than a month
and a half (44 days), her weight came down to 258 pounds (117.3 kUo
grams) andher HgbAlc dropped to 5.2 percent. Thus shelostanaver
age of nearly 3A pound per day and her three-month moving average
blood sugars dropped from 176 mg/dl to 108 mg/dl. Her highest
blood sugar of the final week of this period was 95 mg/dl. Since the
first dose of the new medication, she has been able to foUow our low-
carbohydratemeal plan without hunger, cravings, or any snacking.
The recentlydevelopedincretin mimetics faU into three categories:
Amylin analogs. The onlyone beingmarketedin late 2006is pram-
lintideacetate, brand name Symlin. It ismarketedbyAmyUn Pharma
ceuticals, Inc. It is chemicaUysimUar to amylinand performs the same
functions in the body. It is only avaUable for injection.
GLP-1 mimetics. GLP-1 is one of the hormones secreted into the
bloodstream by the intestines that teU the beta ceUs of the pancreas to
secrete amylin, insuUn, and glucagon. GLP is an abbreviation for
"glucagon-like peptide."The only version currently on the market is
exenatide (Byetta), jointly marketed byAmyUn Pharmaceuticals, Inc.,
and EU LUly and Company. It too is only avaUable for injection.
DPP-4 inhibitors. DPP-4 is an abbreviation for dipeptidyl pepti
dase-IV,the enzyme that the body uses to destroy GLP-1. Administra
tion of its inhibitor opposes the destruction of naturaUy produced
GLP-1. This circumvents the need for injecting a GLP-1 mimetic.
Merckand Company is nowsellinga tablet that can be taken oraUy—
sitagliptinphosphate (brand name Januvia). I question whether it wUl
be as effective for diabetics (whomakelittleor no amylin) as for non
diabetics. I am currently testing it for lowering blood sugars after
The only classof IMs that wiU be effective for curbing the appetite of
people who havevirtuaUy no beta ceUs (type 1diabetics) is the amyUn
206 Treatment
analogs (i.e., pramlintide), sincethe other agentsonly serveto teU ex
isting beta ceUs to make more amylin. Therefore it makes sense to use
only this class forthose of us who make no insuUn.
There is one catch, however. Most type Is havehad high blood sug
ars for more than five years and are therefore likely to have at least
some degree of gastroparesis, ordelayed stomach-emptying. Since one
of the actionsof amylin (andtherefore the other incretinmimetics) is
to slow stomach-emptying, gastroparesis is likely to worsen. As ex
plainedin Chapter22,severe gastroparesis canmakeblood sugar con
trol impossible. If I am faced with a type 1 diabetic who has mUd to
moderate gastroparesis but who also snacks on carbohydrate or
overeats, I must then decidewhich wiU disturb his blood sugarsmore,
the eatingbehavioror the gastroparesis. This canbe a tough caU, but
the decisioncanbe facUitated with the helpof the R-Rstudy described
in that chapter. Any physician contemplating this problemshould cer
tainly readthat chapter.
I have one very obese patient who only overeats at restaurants and
parties. He has moderate gastroparesis. I therefore have him taking
Symlin only beforeeating out. Fortunately, this strategy is working.
VirtuaUy any prescription medication has a potential for adverse
side effects, and the IMs areno exception. Since they do slow stomach-
emptying, their most common adverse effectis gastrointestinal distur
bances such asnausea, constipation, stomachaches, and even diarrhea.
It is therefore wise to start aU ofthem at alowdose and work up slowly
if necessary.
Manufacturers of IMs currently on the market saythey must be in
jectedabout 1hour beforebreakfast and supper. I suggest that excep
tions shouldbe made for those who overeat only between supper and
bedtime, or only at supper, andsoon. In suchcases, I prescribe the IM
about 1 hour before the overeating or snacking usuaUy occurs. For
those who only snack in the lateafternoon and otherwise stick to our
meal plan,I'd prescribe it for useabout 1hour beforethe usualtime of
afternoon snacking. Strangely, some users find that pramlintide must
be injected2-3 hours before the targeted time for it to be effective.
Thereforetiming of injectionsis amatter of trial and error.
It is likelythat in the near future long-acting injectable IMs wUl be
avaUable that need be taken only once weekly.
RecaU from page 98that amylinalso reduces the body's production
of or sensitivity to glucagon — the major culprit in the Chinese
restaurant effect. Between the reduced overeating and the reduced
Howto CurbCarbohydrate Cravingor Overeating 207
glucagon effect, blood sugars can be much lower after meals — even
dangerously lowif you takeblood sugar-lowering medications of any
kind. It is therefore necessary that your physician loweryour doses of
these agents at the time you start any incretin mimetic.
Howmuch should doses be lowered? To some extent it comes down
to experimentation. I usuaUy start by lowering premeal medications
by about 20 percent. I then maylook at blood sugarsthe next day to
see if dosing of these medications should be increased or decreased.
Adjusting Doses of Pramlintide (Symlin)
Pramlintide is suppUed in rubber-stoppered vials, like insulin, but
onlyhalf the size(5 ml). EachmiUUiter contains600megof the drug.
It is injectedwith a standard insulinsyringe, so 1"unit" on the syringe
contains 6 meg. I usuaUy start patients who weigh less than 150
pounds on 2-4 units taken 1 hour before their usual episodes of
overeating or snacking. So if someone only overeats at supper, she
would inject about 1 hour before supper. If he snacks between 9 p.m.
and midnight, he'd inject at 8 p.m. and, if necessary, againat 10p.m. If
someone overeatsonly when eating out, he'd inject about 1 hour be
fore he anticipatesarrivingat the restaurant and not on other days. If
she snacks aU day long, she might inject on arising and every 3-4
hours thereafter.
If someone gets adverse side effects, we'd cut back to a lower dose
until side effects diminish, and then increase each dose lh unit per
week until either cravings vanishor sideeffects reappear. If no adverse
effects appear initiaUy, doses might each be increased by 2 or 4 units
until cravings cease. Heavier people might start at higher doses with
larger increases.
If dosestotal 120units over the courseof a daywithout a major ef
fect on appetite, I would assume that the medication is ineffective.
Adjusting Doses of Exenatide (Byetta)
Until wehave a once-weekly doseof Byetta, it wiU be necessary to fo
cus dosingon the times of daywhen overeating or snacking occurs.
This means starting with the 5 meg prefiUed syringe (orange label)
and injectingabout 1hour beforetimes of snackingor overeating oc
cur. Studythe packageinsert carefuUy to learn howto inject with this
special device,which the manufacturer refersto as a pen. (Avideo tu
torial is avaUable on the manufacturer'sWeb site,www.byetta.com.)
If you overeat only at one meal, inject about 1 hour before that
208 Treatment
meal. If 5 meg helps partiaUy, the dose can be increased to 10 meg by
either injectingtwo 5 meg doses or 1 dose from the 10meg (blue la
bel) pen. If the eating problemoccurs several times daUy, 5-10 meg
should be injected1hour prior to eachtime of the daywhen the prob
lem eating usuaUy occurs. The maximumdaUy dosageof exenatide is
20 meg.
If we ever get a once-weekly dose of exenatide or if Januvia helps
overeating, I wiU certainlyconsiderthem because of their easeof use
(assuming, also, that they are as effective). In the meanwhUe, I prefer
the pramlintide (SymUn) because the amounts injected can be ad
justed in smaU increments and multiple daUy doses are faciUtated,
even though that is contrary to instructions on the package insert.
Also, the pramlintide is much more likely to work for people with
minimal or absent beta ceUs.
Afterreadingthe draft of the aboveparagraph, I haveprescribedfor
a fewpatients a newproduct that stimulates the intestines to secrete
two hormones, GLP-1 and cholecystokinin (CCK). We've already dis
cussed GLP-1. CCK, like amylin, works directly upon the brain to
bring about satiety. Although this newproduct isdouble-barreled,it is
not as effective as injected pramlintide. The new product contains
pinolenic acid, a polyunsaturated fatty acid that's derived from pine
nuts. It is sold as a supplement caUed Natural Appetite Control, by the
Life Extension Foundation. It is worth trying for people who distrust
products approved bythe FDA, but sofar I'veseenno appetitereduc
tions after minimal trials. Dosing is 1-3 softgels about 30minutes be
fore times of anticipated overeating or snacking. The cost is $28 per
bottle if purchased singly, $19 per bottle if purchased in groups of
four. Toorder, phone (800) 544-4440 or visit www.lifeextension.com.
Of all the medications mentioned in this chapter,the IMs certainly
work for more people than the others. If it turns out that they remain
effective after years of use by the same individual, I wiU certainly caU
them miracle drugs.
Recent studies in animals have shown IMs to foster beta ceU regen
eration. I'm sure that this won't cure diabetes, but it may further en
hance the abUity of diabeticsto control and normalize blood sugars.
HowtoCurb Carbohydrate Craving or Overeating 209
1 have seen patients using autohypnosis, low-dose naltrexone, and
evenjust low-carbohydrate dietswho, after one year, found that they
no longer had cravings for carbohydrates or excess food, even when
they discontinued hypnosis or naltrexone. This is probably akin to
some studies of depressed patients treated with certain antidepres
sants: after a period of time, the antidepressants may no longer be
needed.MetaboUc brain scans sometimes showan apparentlyperma
nent normalization of brain function in selected regions. For aU we
know, this mayapplyas weU to the incretinmimetics.
One of my pramlintide patientsfound that it ceased workingafter
2 months of use. She discontinued it for one week and restarted at a
lower dose. The lowerdoseis stillworking after2V2 months.
In any event, the improvement in blood sugars and concomitant
weightlossare major incentives to give thesemethods a try.
Some Symlin users have complained that the stoppers on some of
theirSymlin vials have beensohardthat the needle of their insuUn sy
ringe is duUed after one fiUing with Symlin. They also have com
plained that on some vials, the needle actuaUy becomes permanently
bent when one attempts to puncture the vial.
Amylin Pharmaceuticals is planningto comeout with an improved
stopper,but approvalof this product bythe FDAmaydelaythe release
by several years from the time of this writing. The company is also
planning to come out with a Symlin pen simUar to the pen used for
Byetta. This is not reaUy a solution to the problem, sincethe pen wiU
only inject fixeddoses instead of the variable doses that we seek.
Thisleaves uswithtwooptionsfor circumventing the problem.The
first is to puncture the stopper in the verycenter, where the rubber is
thinnest. The needle should be perpendicular to the surface of the
stopper and should not puncture it at an angle. If this technique does
not solve the aforementioned problems, then it wiU be necessary to
210 Treatment
transfer your Symlin to empty vials that aremore suited to use with
insulin syringes.
Empty, sterUe insulin vials are avaUable at no charge from most
pharmacies. They are also avaUable fromEU LUly and Company.Your
physician can prescribe theseempty vialstogetherwith a 5 cc syringe
with a IVi-inch, 23 gauge needle for transferring the Symlin from the
original vial to anew 10ccinsulinvial. The technique fortransferring
the SymUn is very simple. Puncture an empty vial with the 23 gauge
needle and drawout 5 cc of air. Inject half the airinto the SymUnvial.
With the vial upside down andthe needleweUbelowthe surfaceof the
Uquid, draw out about 2.5ccof SymUn. Then injectthe rest of the air
into the Symlin vial and drawout the remaining2.5cc of Symlin.
You now can inject the 5 cc of Symlin into the empty, sterUe vial. If
you wish, you canrepeat the above procedure with asecond5 ccSym
Unvial, transferring its contents to the same empty 10 cc insuUn vial
that you used previously. Youcan, forexample,transferthe contents of
ten Symlin vialsto five 10ccinsulinvials.
Anyone can purchase a subscription to Obesity Meds and Research
News at www.obesity-news.com, or by phoning (703) 960-5513 or
faxing (703) 960-7462. This is the best update sourcethat I know of
for news on upcoming and newly approvedmedications.
Using Exercise to Enhance
Insulin Sensitivity
Strenuous, prolonged exercise is the next level of our treatment
plan after diet, and should ideaUy accompanyany weight-loss
programor treatment for insulin resistance (as in type 2 dia
betes). Before we go into our specific recommendations for exercise,
aU of which should be approved by your physician prior to putting
them into practice,it's important that you understand the benefits ex
ercisecan bring.
WhUe many peoplemaybeginexercising out of asenseof responsibU-
ity — the way chUdreneat vegetablesthey don't like — the main rea
son they keep exercisingis that it feels good. Whether it's the intense
competition of a fast and furious basketbaU game, or cyclingalone in
the countryside,exercise bringsmany rewards — physical, psycholog
ical, and social.
People who aren't diabetic and exercise strenuously and regularly
tend to Uve longer, are healthier, look healthier and younger, have
lower rates of debUitating and incapacitating Ulnesses such as osteo
porosis, heart disease, high blood pressure, memory loss of aging—
and the Ust goes on. OveraU, peoplewho exercise regularly arebetter
equipped to carryon day-to-day activities as they age.
Many type 1diabeticshavebeen Ul for so long with the debUitating
effects of roUer-coaster blood sugars that they are often depressed
about their physical health. Numerous studies have established a link
between good health and a positivemental attitude. If you'rea type 1
212 Treatment
diabetic,asI am, strenuous exercise wiUnot improve your blood sugar
control as it wiU for type 2s (which we'U discuss shortly), but it can
have a profound effect on your self-image. It's possible, if you keep
yourbloodsugars normal andexercise regularly andstrenuously, to be
in betterhealththanyournondiabetic friends. Also, it'sbeenmy expe
rience that type 1diabetics who engage in aregular exercise program
tend to takebetter care of theirblood sugars and diet.
Think of exercise as money in the bank — every 30 minutes you
put intokeeping in shape today wUl not onlyleave youbetter off right
now, it wiU paycontinuing dividends in the future. If going up the
stairs yesterdayleft youhuffingandpuffing, in awhUe you'll bound up
the steps. Your strength wUl likely make you feel younger, possibly
more confident. There is evidencethat exercise actuaUy does make you
lookyounger: even the skinof those whoexercise regularly tendsnot
to ageasrapidly.
After workingout for a few months, you'U look better, and people
wUl mention it. With this kind of encouragement, you may be more
likelyto stickto otheraspects of our regimen.
Although most of us who engage in bodybuUding exercise canex
perience increases in muscle mass and strength, the degree to which
we respond is in part geneticaUy determined. With verysimUar exer
cise regimens, somepeople wiU showdramatic increases in both mus
clebulk andstrength; others wUl showneither. Most of usUe between
these two extremes. There areeven peoplewho gain strength but not
large muscles, andothers whocan buildlarge muscles without getting
much stronger. If you don't develop big muscles or great strength,
you wiU stiU enjoy the other benefits from the weight training de
scribed here.
It haslongbeenknown that strenuous exercise raises the levels of
serumHDL (good cholesterol) andlowers triglycerides in the blood
stream. Recent studies suggest that bodybuUding exercise (anaerobic
rather than aerobic exercise) also lowers serum levels of LDL (bad
cholesterol). There is even evidence that atherosclerosis (hardening of
the arteries) may be reversible in some individuals by such major im
provements in serumUpid profiles. I'm weU over seventy, I exercise
strenuously on adaUy basis, I don't eat fruit, I've hadtype 1diabetes for
sixty years, and I have two eggs for breakfast everyday. Where's my
cholesterol? It'sin averyhealthyrange that nondiabeticsone-third my
age rarely attain (page 131). Part of thatisdueto mylow-carbohydrate
diet, but part of it is due to my daUy exercise program.
UsingExercise toEnhance InsulinSensitivity 213
Frequent strenuous exercise hasbeen shown to reduce significantly
the Ukelihood of heartattack, stroke, andblockage of bloodvessels by
lowering serum fibrinogen levels. Long-term strenuous exercise low
ers resting heart rate and blood pressure, further reducing the risk of
heart attacks and stroke.
Weight-bearing, resistance, and impact exercise slow the loss of
bone mineral associated with aging. Ever hear the slogan "Use it or
lose it"? In avery real sense, if we don't use our bones, we lose them.
Although exercise does make weight control easier, it does not di
rectly — atleast not as muchas wemaywish— "burnfat." Unless you
workout at verystrenuous levels for several hours each day, exercise
isn't going to have asignificant direct effect uponyour body fat. The
effects of exercise are broader andmore indirect. Oneof thegreat ben
efits is that many people find that when they exercise, they have less
desire to overeat. The reasons for this are probably related to the re
lease in the brain of neurotransmitters suchas endorphins. (As noted
in the previous chapter, endorphins are "endogenous opiates" manu
factured in the brain. They can elevate mood, reduce pain, andreduce
carbohydrate craving. Brain levels of endorphins are reduced in poorly
controUed diabetes.*) It might be said that in the samewaythat obe
sity leads to further obesity, fitness leads to further fitness.
Even though your fat won't "melt away," exercise, particularly if
you're atype 2diabetic, isstill of value inaweight-reduction program
because muscle buUding reduces insulin resistance. Insulin resistance,
remember, is linked to your ratio of abdominal fat to lean body
mass.The higher your ratioofabdominal fatto muscle mass, the more
insulin-resistant you're likely to be.Asyouincrease yourmusclemass,
your insuUn needswiU be reduced — andhaving less insulin present
in your bloodstream wUl reduce the amount of fat you packaway. If
you remember my old friend Howie from Chapter 12, his insulin re
sistance dropped dramatically when helost 100 pounds andradicaUy
changed his ratio of abdominal fat to lean body mass.
Long-term, regular, strenuous exercise also reduces insulin resis
tance independently of its effect upon muscle mass. This makes you
more sensitive to your own and injected insulin. As a result, your in-
*Plasma endorphin levels can be measured bymost commercial laboratories. If
youarecuriousabout your level, just askyour physician to order plasmabetaen
dorphin prior to startingan exercise program or a regimen of naltrexone and
againa fewweeksor months later.
214 Treatment
sulin graduaUy becomes more effective at lowering blood sugar. If
you inject insulin, your dosage requirements wiU drop, and the fat-
buUding effects of largeamounts of insulin wiU likewise drop. In my
experience, daUystrenuousexercise wiU, overtime,bringabout a steady,
increased level of insuUn sensitivity. This effect continues for about
two weeksafter stopping an exercise program. Awareness of this is es
peciaUy important for those of us who inject insulin and must in
creaseour doses after two weeks without our usual exercise. If you go
out of town for only a week and cannot exercise, your increasedin
sulin sensitivity wiU probablynot suffer.
Although increased muscle bulk also increases insulin sensitivity,
independently of the aboveeffect, this isverygradual and mayrequire
many months of bodybuUding beforeits separateblood sugar effects
become noticeable.
Exercise does affect blood sugar, and for that reason it can make your
efforts at blood sugar control sUghtlymore difficult if you're taking in
sulin or sulfonylurea blood sugar-lowering medications.* The bene
fits, however, are so great that if you're a type 2 diabetic, you'd be
fooUsh not to get involved in an exercise program.
For years, guidelines for the treatment of diabetes haverepeatedthe
half-truth that exercise always lowers blood sugar levels. In reaUty,
physical exertion can indeedlowerblood sugar via increasednumber
and mobilization of glucosetransporters in muscle ceUs. Certain con
ditions, however, must be present: exertion must be adequately pro
longed, serum insulin levels must be adequate, blood sugar must not
be too high, and for most of us, exercise should not be performed
within 3 hours of arisingin the morning (seepage216).
Moderate to strenuous exercise such asswimming,running, weight
Ufting, or tennis — as opposed to more casual exercise, such as walk
ing — causes an immediate release of "stress," or counterregulatory,
hormones (epinephrine, Cortisol, et cetera). These signal the Uver and
*As I will explainin Chapter 15,1 recommendagainst the use of sulfonylureas
and similar medications.
UsingExercise to Enhance Insulin Sensitivity 215
muscles to return glucose to the bloodstream by converting stored
glycogeninto glucose.The nondiabetic responseto the additional glu
cose is to release smaU amounts of stored insulin to keep blood sugars
from rising. Blood sugarthereforewUl not increase. If atype 2 diabetic
without phaseI insulin responsewereto exercise fora fewminutes, his
blood sugar might increase for awhUe, but eventuaUy it would return
to normal, thanks to phase II insulin response. Thus, brief strenuous
exercise can raise blood sugar, whileprolonged exercise canlower it. For
this reason, Dr. EUiot P. Joslin told a group of us (in 1947): "Don't run
a block for a bus, run a mUe."
When insulin is nearly absent in the blood, the glucosereleased in
response to stress hormones cannot readUy enter muscle and Uver
cells. As a result, blood sugar continues to rise, and the muscles must
rely upon stored fat for energy. On the other hand, suppose that you
have injected just enough long-acting insulin within the previous 12
hours to keep your blood sugar on target without exercise, and then
you run a few mUes. You wiU have a higher serum insuUn level than
needed, because exercise facUitates the action of the insulin already
present. Blood sugarmay thereforedrop too low. The same effect may
occur if you are using sulfonylureas, a class of oral hypoglycemic
agents. Furthermore, if you haveinjectedinsulin into tissue that over-
Ues the muscle being exercised, or perhaps into the muscle itself, the
rate of release of insulin into the bloodstreammay be so great as to
cause serious hypoglycemia. Nondiabetics and type 2s not on insulin
or sulfonylureas can automaticaUyturn down their insuUnin response
to exercise.
It may be unwise for you to exercise if your blood sugar exceeds
about 170 mg/dl. This number varies with the individual and the
medications taken. This is because elevated blood sugars wiU tend to
rise even further with exercise. This effect wiU be less dramatic if
you're making a lot of insulin, and is most dramatic for a type 1 dia
betic who doesn't take extra insulin to prevent the blood sugareleva
tion. I have one type 1 patient who keeps her blood sugars essentiaUy
normal.ShestUl makesalittleinsulinanddislikes insulininjectionsso
much that sheworks out everydayafterlunch to save herself ashot to
cover the lunch. In her case, the exercise plus the smaU amount of in
sulin she stiU makes together work very weU.
One greatbenefit of regular, strenuous exercise in type 2 diabetes, as
mentioned earlier, is that it canbring about a long-term reduction of
insulin resistance, by increasing muscle mass. Long-term muscle de-
216 Treatment
velopment, therefore, can faciUtate blood sugar control and weight
loss. It also reduces the rate ofbeta ceUburnout, because the increased
ratio of muscle mass to abdominal fat reduces insulin resistance and
thus reduces the demand for insulin production.
Several of my type 1 patients must take additional fast-actinginsulin
when they exercise in the morning, but not when they exercise in the
afternoon. This is a dramatic exampleof howthe dawn phenomenon
reduces even injected serum insulin levels. In the afternoon their
blood sugardrops withexercise, but in the morning it actuaUy goesup
if they do not first inject some rapid-acting insuUn.
Despite the benefits that exercise can have, an exercise program that
isn't sensibly put togethercanhave disastrous results. Even if you think
you're perfectiyfit, your physicianshould be consulted before you pro
ceed. Keep in mind that there are certain physical conditions that may
restrict the type and intensityof exercise you should attempt. Yourcur
rent age, your cardiac and muscle fitness, the number of years you've
had diabetes, the average level of your blood sugars,whether or not —
and howmuch — you'reoverweight, and what sort of diabetic compli
cationsyou havedeveloped: aU thesemust be considered to determine
what kind of exercise youshould undertake,and at what intensity.
Before You Start
FoUowing are severaldifferent aspectsof your health you should con
sider and discusswith your physicianbeforeembarking upon an exer
cise program.
Heart. Everyone over the ageof forty, and diabetics over the age of
thirty, should be tested for significant coronary artery disease before
beginning a newexercise program. At the veryleast, an exercisingelec
trocardiogram, stress echocardiogram(my preference), or stress thal
lium scan is usuaUy advised. An abnormal test may not necessarUy
Using Exercise toEnhance Insulin Sensitivity 217
rule out exercise, but it may suggest restraint or close supervision
whUe exercising. Again, seek your doctor's advice before starting any
new exercise program.
High blood pressure. Although long-term exercise helps to lower
resting bloodpressure, your bloodpressure canrisewhUe you areex
ercising. If you're subject to wide pressure swings, theremaybe riskof
stroke andretinal hemorrhages during strenuous exercise. Again, first
contact your physician.
Eyes. Before beginning anyexercise program, youshouldhave your
eyes checked bya physician, ophthalmologist, or,ideaUy, a retinologist
experienced in evaluating diabetic retinal disease (retinopathy). Cer
tain typesof retinopathy arecharacterized bythe presence of neovas
cularization, or very fragUe newbloodvessels growing from the retina
into the vitreous gel that overUes it. If youstrain too much, assume a
head-down position, or landhardonyourfeet, these canruptureand
hemorrhage, causing blindness. If yourphysician or ophthalmologist
identifies suchvessels, you'U probablybewarnedto avoidexercises re
quiringexertion of strongforces (e.g., weight Ufting, chinning, push
ups, sit-ups) and sudden changes of motion (e.g., running, jumping,
falling, diving). Bicycling and surface swimming are usuaUy accept
able alternatives, but first check withyourphysician.
Fainting. A form of nerve damage caUed vascular autonomic neu
ropathy (caused by chronically highblood sugars) can lead to light
headedness and even fainting during certain types of exertion (see
page 341), such as weight Ufting and sit-ups. Such activities should
therefore be embarked upon graduaUy and onlyafter instruction by
your physician.
If you take blood sugar-lowering medications. If you take in
sulin or oral hypoglycemic agents, it is wise to makesure your blood
sugars are stabilized beforeyou begin a strenuous exercise program.
As previously noted, exercise can have significant effects upon blood
sugars and introduce another variable that can confuse anyone re
viewing your blood sugar data. It's much easier to readjust your diet
and/or medications to accommodate physical activityafter bloodsug
ars are under control.
218 Treatment
Sympathetic autonomic neuropathy. If you're unable to sweat be
lowyour waist, thereisa possibUity that prolonged exercise maycause
undue elevationof your body temperature.
Proteinuria. Elevated levels of urinary protein are usuaUy exacer
bated by strenuous exercise. This in turn can accelerate the kidney
damage that youmayalready have.
Ongoing Concerns for Exercising Diabetics
FoUowing is a Ust of aspects of healthyoushouldconsider on an on
goingbasis asyoupursueyour exercise program.
Recent surgery. A history of recent surgery usuaUy warrants re
straint or abstinence until youreceive clearance fromyour surgeon.
Blood sugar changes. Even after blood sugars are reasonablyweU
controUed, illness, dehydration, and eventransient blood sugar values
over 170 mg/dl arereasons for youto refrain fromexercise. For many
people, bloodsugars above 170 mg/dl wiU increase further withexer
cise, due to the production of stress hormones that wediscussed pre
Blood sugars belowtarget values. Ifyoutakebloodsugar-lowering
medications, do not exercise if blood sugar is belowyour target value.
Bring it up totargetfirst withglucose (see thenextsection, and Chap
ter 20, "How to Prevent and Correct LowBlood Sugars").
Possible foot injury. If you've had diabetes for a number of years,
thereisagoodchance that yourfeet areespeciaUysusceptible to injury
whUe exercising. Thereare several reasons for this:
• The circulation to your feet maybe impaired. With a poor blood
supply, the skin is readUy damaged and heals poorly. It also is
more likely to be injuredbyfreezing temperatures.
• Injuryto nerves in thefeet caused bychronicaUy highbloodsug
ars leadsto sensoryneuropathy, or diminishedabUity to perceive
pain, pressure, heat, cold, and so on. This enables blisters, abra
sions, and the like to occur and continue without pain.
• The skin of the feet can become dry and cracked from another
form of neuropathy that prevents sweating. Cracks in heels are
potential sitesof ulcers.
Using Exercise to Enhance Insulin Sensitivity 219
• A third form of neuropathy, caUed motor neuropathy, leads to
wasting of certain muscles in the feet. The imbalance between
stronger and weakermuscles leadsto a foot deformityverycom
mon among diabetics, which includes flexed or claw-shaped
toes, high arches, and bumps on the sole of the foot due to
prominence of the heads of the long metatarsal bones that lead
to the toes.Theseprominent metatarsal headsare subject to high
pressure during certain types of weight-bearing exercise. This
can lead to caUuses and even skin breakdown or ulcers. The
knuckles of the claw-shaped toesaresubjectto pressurefromthe
upper surface of your shoes or sneakers. The overlying skin can
therefore blister and ulcerate.
• Another form of neuropathy makes it difficult to perceive joint
positionin the feet. This, in turn, canleadto orthopedicinjuries
(e.g., bone fractures) whUe running,jogging, or jumping.
AU of this implies that the feet must be carefuUy protected during
exercise. Your physician or podiatrist shouldbe consulted before you
start any new exercise, as some restrictions may be necessary. Even
prolonged swimming can cause maceration of the skin. You should
also be thoroughlytrained in foot care. Please seeAppendix D,"Foot
Care for Diabetics."
You or a famUy member should examine your feet daUy for any
changes, abrasions, pressurepoints, pink spots, blisters, and so on. Be
sureto checkthe soles of your feet, usinga hand mirror if necessary. If
youfindanychanges, seeyour physician immediately. Bring withyou
aU the shoes and sneakers that youcurrently use, so that he can track
down the cause of the problem. At the very least he mayrecommend
the use of flexible orthotic inserts and wide, deep toe box sneakers
whUe exercising. Noattempt shouldever be madebyanyone (includ
ing podiatrists) to remove calluses, as this is probably the most com
mon cause of ulceration.
People who do not take medications that lower bloodsugar are usu
aUy ableto "turn off" their insulin secretion in response to a drop in
220 Treatment
blood sugar brought about by exercising. You cannot, however, turn
off sulfonylureahypoglycemic agents or injected insulin once you've
taken them. (This is one of the reasons I never prescribe sulfonylureas
and simUar products.) To prevent the occurrenceof dangerously low
blood sugars, it is wise to coverthe exercise with glucose tablets (e.g.,
Dextrotabs; see page326) in advance of a drop in blood sugar.
Some type 1 diabetics try to use"treats," such as fruit or candy, to
cover an anticipatedblood sugar drop. I don't ordinarilyrecommend
this approach, becauseit's not as precise as using glucose tablets. My
experience with patients who've taken raisins or grapes or candies to
cover their exercise has been that they suffer subsequent elevated
blood sugars. Say you eat an apple. It wiU contain some fast-acting
sugars that enterthe bloodstream almostimmediately. It wiU also con
tain other, slower-acting sugars that may take several hours to have
their fuU effect upon blood sugar. On the other hand, as we wiU dis
cussbelow, certainsustainedactivities — such ascross-country skung
or physical labor for many hours — cankeep your blood sugar drop
ping aU day. For those, you'll need something longer-acting to help
keep you frombecominghypoglycemic.
To discover howmuch carbohydrate you shouldtake foragivenex
ercise session requires some experimentation and the help of your
blood sugar meter. One valuable guideUne is that 1 gramof carbohy
drate wUl raise blood sugar about 5 mg/dl for people with body
weights in the range of 140 pounds. A chUd weighing 70 pounds
would experience double the increase, or 10 mg/dl per gram, and an
adult weighing 280 poundswould probably experience only half this
increase (2.5 mg/dl).
My own preference is Dextrotabs, each of whichcontains1.6grams
of glucose. Other brandsof glucose tablets are avaUable at most large
pharmacies anddiabetes maU-order suppUers (see Chapters 3and20).
If you weigh 150 pounds, one Dextrotab wiU raise your blood sugar
about 8 mg/dl. Sincethese glucose tablets start raising blood sugar in
about 3 minutes and finish in about 40minutes, they'reideal for rela
tively brief exercise periods.
Let's run through a hypothetical example to demonstrate how
you'd goabout determininghowmany tabletsyou ought to take. Let's
assume you weigh 170 pounds and 1 Dextrotab wiU likely raiseyour
blood sugarabout 7 mg/dl. You'vedecidedto swim (or playtennis) for
an hour.
UsingExercise to Enhance Insulin Sensitivity 221
• First, check yourbloodsugar before starting (youshould always
check blood sugar before starting to exercise). If it's belowyour
target value,takeenoughDextrotabs to bringit up to target. Wait
40 minutes for them to finish working. If you don't come up to
your target, you may be too weak to exercise effectively. Record
your blood sugar level upon starting. (I urgethe use of Gluco
graf data sheets for recording aU exercise-related blood sugars.)
• When you beginsuchactivity — the first time you exercise after
beginning our regimen — take 1 Dextrotab, and then 1 again
every 15 minutes thereafter.
• Halfway into your activity, checkblood sugar again, just to make
sureit's not too low. If it is, take enough Dextrotabs to bring it
backup, andcontinuethe exercise. If it'stoo high,you mayneed
to skipthe next few tablets, depending upon howhighthe value.
• Continue the exercise and the tablets (depending upon blood
• At theendof the exercise period, measure bloodsugar again. Cor
rect it with glucose tablets if necessary. Remember to write down
aU blood sugarvaluesand the time when eachtablet was taken.
• About an hour after finishing your workout, checkblood sugar
again. This is necessary because it may continue to drop for at
least 1hour after finishing. Bring it backup with glucose tablets
if necessary. (Very intense or prolonged exercise maykeepblood
sugarsdropping for as long as 6 hours.)
• If you required, say, a total of 8 tablets altogether, this suggests
that in the future you should take 8 tablets spread out over the
course of yourworkout. If youonlyrequired 4tablets, then you'd
take 4 tablets the next time. And so on. For some exercise pro
grams you may need no tablets.
• Repeat thisexperiment on occasion, because youractivitylevel is
rarely exactly the same for every exercise period. If you required
3 tablets the first time and 5 tablets the second time, take the
average, or 4 tablets, the next time. If your activity level in
creases — sayyou'vebeen playing with aslowtennis partnerand
you find another whomakes yousweat yourbutt off—youmay
find it necessary to increase the number of glucose tablets.
Thereare someactivities where coverage with aslower-acting form
of carbohydrate may be appropriate, andit's here, perhaps, that you
222 Treatment
could use the "treats" I would normally discourage. For example, I
have two patients, both on insulin, who are housepainters. Neither
works every day, and the hours of work vary from day to day. They
rarelyworkfor less than 4 hours at a time.The painter in Massachu
settsfinds that half a blueberrymuffineveryhour keepshis blood sug
ars level, whUe the painter in NewYork eats a chocolate chip cookie
Somepatients find that their blood sugars drop when they spend a
fewhours in a shopping maU. I teU them to eat a sUce of bread (12
gramscarbohydrate) whentheyleave their car. ThebreadwiU start to
raiseblood sugar in about 10minutes, and wUl continue to do so for
about 3hours. The cookies and blueberrymuffinscontain mixtures of
simpleand complexsugars, sotheystart working rapidlybut alsocon
tinue to raiseblood sugar for about 3 hours. I discourage the use of
fruits, which can raise blood sugar lesspredictably. If your exercise is
not goingto continue for manyhours, coverit with glucose — not a
fun food — if you want predictableresults.
Beware, however. If you havea history of cravingcarbohydrate, fun
foods are likely to exacerbate the problem, making the addiction im
possibleto control.
Whatever your plan for coveringexercise with carbohydrate, always
carry glucose tablets with you! If you have gastroparesis, you may do
better with a Uquidglucosesolution (seepage 377).
As you are by nowaware, insulin resistance, which is the hallmarkof
type 2 diabetes, is enhanced in proportion to the ratio of abdominal
fat to lean body mass. One of the best ways to improve this ratio in or
der to loweryour insuUn resistance is to increase your leanbody mass.
Therefore, for most type2diabetics, the most valuable type of exercise
is muscle-buUding exercise. (It'sgoodfor type Is too, becauseit makes
you feel better, look better, and can improve your self-image.) There
also is cardiovascular exercise, which benefits the heart and circulatory
system, and wiU be discussed later in the chapter.
First, what is muscle-buUding exercise? Resistance training, weight
training (weight lifting), or gymnastics would aU qualify. If done
properly, weight liftinghas many attributes that make it superior to
Using Exercise to Enhance Insulin Sensitivity 223
theso-caUed aerobic exercises. Aerobic exercise isexercise mUd enough
that your muscles are not deprived of oxygen. When muscles exercise
aerobicaUy, theydon't increase muchin mass andtheydon't requireas
much glucose for energy. Anaerobic exercise deprives the muscles of
oxygen; it tiresthemquicklyand requires nineteentimesas much glu
cose to do the same amount of work as aerobic exercise. When you
perform anaerobic exercise, your muscles break down for the first 24
hours, but then they buUdup over the next 24 hours. I havelittle old
ladies performing weight-Ufting exercise. They're never going to look
like Arnold Schwarzenegger —it's physicaUy impossible because
women haven't the hormonesfor it —but theyfeel much better and
arecertainly stronger because of it. Theyalso buUd enoughmuscleto
reduce their insulin resistance.
Butwhat about aerobic exercise, suchasjogging or outdoor biking?
I don't think it's as valuable for diabetics —or for anyone reaUy, for
reasons weshaUdiscuss. StiU, I usuaUy suggest that mypatientsengage
in activities that theywiU enjoyand wiU continueto pursue in a pro
gressive fashion. Progressive exercise is exercise that intensifies over a
period of weeks, months, or years. Beloware Usted various character
isticsof an appropriate exercise program:
• It should comply with any restrictions imposed by your physi
• The cost should not exceed your financial limitations.
• It should maintain your interest, so that you'U continue to pur
sue it indefinitely.
• Thelocation shouldbeconvenient, andyoushouldhave thetime
to workout at least every other day. DaUy activity is very desir
• It should be of a progressivenature.
• It shouldideaUy buUd muscle mass, strength, and endurance.
• The same muscle groups should not be exercised anaerobicaUy
two days in a row.
You've often heard of aerobics, and now you've seen me mention
"anaerobic" several times. What makes oneof these types of exercise
better for diabetics than the other?
224 Treatment
Our musclesconsist oflong fibers that shorten, or contract,when they
performwork likeUfting aloador moving the body.AU muscle fibers re
quirehigh-energy compoundsderived from glucose or fatty acids in or
der to contract. Some muscle fibers utilize a process caUed aerobic
metabolism to derive high-energy compounds from smaU amounts of
glucose andlarge amounts of oxygen. These fibers canmove Ught loads
for prolonged periods of time, andare most effective for "aerobic" pur
suits, such as jogging, race walking, aerobic dancing, tennis, nonsprint
swimming, moderate-speedbicycling, andsimUar activities. Other mus
cle fibers canmove heavyloads but only forbrief periods. They demand
energyat avery rapid rate, andso must be able to produce high-energy
compounds faster than the heart can pump blood to deUver oxygen.
They achieve this by a process caUed anaerobic metabolism, which re
quires large amounts of glucose andvirtuaUy no oxygen.
This is of interest to diabetics fortwo reasons. First, the blood sugar
drop during and after nearly continuous anaerobic exercise wiU be
much greater than after asimUar periodof aerobic exercise because of
this requirement for large amounts of glucose. Second, as your body
becomes accustomed to this requirement, it wUl adjust to the stresses
you put on it and more efficientlytransport glucose into your muscle
ceUs. As muscle strength and bulk develop, glucose transporters in
these ceUs wiUincrease greatlyin number. Glucose transporters also
multiply in tissues other than muscle, including the liver. As a result,
the efficiencyof your own (or injected) insulinin transportingglucose
and in suppressing glucose output by the liverbecomes considerably
greater when anaerobic exercise is incorporated into your program.
In relatively short order, you wiU develop greater insulin sensitivity
for lowering blood sugar. SimUarly, your requirements for insulin
(that which you create or inject) wiU diminish.The overaU dropin in
sulinin your bloodstreamwiU reduce yourbody'sabilityto hold on to
stored fat, thus further lowering insulin resistance.
Think here of the Pimas. Not only did they gain access to an almost
unlimited food supply, they also went from a strenuous existence,one
that naturaUy incorporated both aerobic andanaerobic activity, to one
that was almost entirely sedentary. Thus their circumstances were
changed utterly fromwhat you might caU the biological expectations
of their bodies. Of course, it's not just the Pimas who are sedentary.
When you understandhowto meet yourbody'sevolutionaryexpecta
tions, you canbeginto bring it backinto balance.
Anaerobic metabolism producesmetabolic by-products that accu-
Using Exercise toEnhance Insulin Sensitivity 225
mulatein the active muscles, causing pain and transient paralysis —
for a fewseconds, youjust can't contractthat muscle again. Since these
by-products are cleared almost immediatelywhen the muscles relax,
the pain likewise vanishes upon relaxation, as doesthe paralysis. You
can identify anaerobic exercise bythe local pain and the accompany
ingweakness. This painisUmited to themuscles being exercised, goes
away quickly when the activity stops, anddoes not refer to agonizing
musclecramps or to cardiacpain in the chest. Anaerobic activities can
include weight Ufting, sit-ups, chinning, push-ups, runningup a steep
incline, uphitt cycling, gymnastics, using a stair climber, and so forth,
provided that these activities are performed with adequate loads and
at enoughvelocity to cause noncardiac pain or transient discomfort
(not heart attack, but the painof "no pain,no gain").
Continuous anaerobic activity, as you canweU imagine, is reaUy im
possible. The pain in the involved muscles becomes intolerable, and
theweakness that develops with extreme exertion leaves youunable to
continue. Nevertheless, youcanapproach thisgoal byusingthespecial
"inverted pyramid" technique described on page 228.
BodybuUding, or resistance exercise —which includes weight Uft
ing, sit-ups, chinning, and push-ups —focuses on one muscle group
at a timeand then shifts the focus to anothermuscle group.If you use
the inverted pyramid technique you can achieve nearly continuous
anaerobic activity, but on a rotating basis. After you finish exercising
certain of your abdominal muscles bydoing sit-ups, for instance, you
switch to push-ups, which focus on various arm and shoulder mus
cles. Fromthere, yougotochinning. SimUarly, different weight-lifting
exercises also focus on different muscle groups. Anaerobic exercise
also canincrease the benefits of exercise in stimulating heart rateand
thereby exercising the heart. To maintain an elevated heart rate, you
switch immediately from one anaerobic exercise to another, without
resting in between.*
I personaUy preferthis typeof activity for type 2 or obesediabetics
*This type of cardiac exercise is not nearly as effective in raising heart rate as
those described under"Cardiovascular Exercise," page 230.
226 Treatment
because — as I have said before and wiU say again — the buildup of
muscle mass lowers insuUn resistance and thereby facUitates both
blood sugar control and weightloss. Anumber of mypatients engage
in bodybuUding exercises, includingmen and womenover sixtyyears
of age. TheyareaU verypleased withthe results.*
Sincethe pubUcation of the first edition of this book, there has been
a change in our societyin the recognition of the importance of this
kind of exercise. Asignificant benefitis its abUity to helpincreasebone
density. Bones, likemuscles, tend to be onlyas strong as they need to
be. When you strengthen your muscles, you're also exercising your
bones —your muscles, afteraU, areattachedto your bones;whenthey
contract, your bones move on their joints. If your bones weren't as
strongas the muscles attached to them, they'dsnap.
Some Suggestions for a Bodybuilding Routine
Please refer back to "Restrictions on Exercise," page 216. These restric
tions and cautions applyespeciaUy to bodybuUding.
Even if you have room in your home, and the finances, to equip
your ownprivate gym, I usuaUy recommend that people goto an out
sidegymor healthclubto learn the different exercises before begin
ning an anaerobic exercise program. Then, if you want to buy
dumbbeUs or a weight-lifting machinefor useat home, that's fine. But
it's important to learn good technique and good form first. You can
also consult books on the subject, but at least a fewsessionssupervised
by an experiencedinstructor is best.
Equipment. Foryour upperbody, you're goingto haveto useweights.
I don't recommend that you Uft barbells— they can be dangerous,
and you therefore must have assistance if you're using them — but I
do recommend dumbbeUs and weight-Ufting machines, which for the
most part are quite safe to use.* Whether you're using dumbbeUs at
*Anumber of years ago, a report fromthe human physiology lab at Tufts Uni
versity reported that onlytwelve weeks of weight trainingtripledthe strength of
malesubjects ages 60-96. Thiswas believed to improve their qualityof life sig
nificandy. Subsequent studiesshowed similareffects in women.
f Anumber of inexpensive multiexercise machinesare on the market that utilize
thick rubber bands instead of weights. Beware of these: since you have to stop
and change the bands to change resistance, they do not permit true anaerobic
training. Hydraulic and pneumaticmachines that utilize rotary knobs to adjust
settingsare usuallyexcellent but oftenquite cosdy.
Using Exercise to Enhance Insulin Sensitivity 227
homeor in the gym, they should be soUd cast iron, usuaUy painted
black enamel or gray. They're inexpensive —usuaUy 50-75 cents a
pound, so a 10-pound dumbbeU costs about $5-$7.50. Don't use
dumbbeUs consisting of a bar with plates on either end that can be
added or removed. These can be dangerous —the plates frequently
sUde off— and theyalsodefeat the wholemethod that I advocate (see
"Technique," below).
Exercises. Ifyou're going toa health club or gym tolearn theropes,
I suggest that you learn fifteen upper body exercises, and as many
lower bodyexercises as are avaUable. Upper bodywould be for the
arms, hands, shoulders, flanks, chest, abdomen, andback. Ifyou're go
ingto the gym every day, which I recommend, you'd do your upper
body exercises on oneday, andlower body exercises the next. Why al
ternate days? Becauseof the muscle breakdown over the first 24 hours
after exercise andthe need for time to rebuUd. So on the second day,
whUe you're doing your lower body exercises, your upper body mus
cles are rebuUding.
As you can guess, there are more muscle groupsthat workin more
ways in the upper bodythan in the lowerbody, so there are fewer sen
sible lower bodyexercises. If you're using a treadmiU, a stair, a bike,
anda cross-country ski machine aU inthesame day, you're exercising
more or less thesame lower body muscles witheach apparatus, which
isn't sensible. The other types of lower body exercises that involve
weight lifting are few in number: leg presses, knee curls, toe presses,
andknee extensions. Some gyms have machines to exercise yourlegs
asyouspreadthemapart or squeeze themtogether, but in aU, thereare
at most sixlegexercises commonlyavaUable.
Asa consequence, I always add some other exercises on the days I
do lower bodyexercises: gripstrengthening, side bends (which exer
cise the sidemuscles), andsit-ups or crunches, asweU aswhat's caUed
cardiovascular exercise (page 230). Theinstructor at yourhealth club
wiU be able to help you with aU of these.
Form. To get the most out of your weight-Ufting exercises, it's im
portant to have as close to perfect formas possible. This means that
youisolate and useonlythe muscles targeted bya particular exercise.
You shouldn't, for example, use yourback muscles to help perform an
arm exercise. You should also Uft slowly, saygraduaUy over about 7
seconds, andlet theweight downveryslowlyoverabout 15seconds, so
228 Treatment
that the entire individual repetition takes about 25 seconds — or as
longasyoucantolerate. This tends tobemucheasier onjointsandhas
been shownto be a higher quahtyof exercise. Do not fuUy flex or ex
tend your muscles while weight lifting. Instead, stop just before you
would reach the endpoint of any motion. This is where having good
instruction can pay off. Your instructor can critique your form and
helpyouselect the right equipmentfor each exercise.
Technique: The Inverted Pyramid System
First, a word of warning: Don't embark upon this technique until you
have demonstrated perfect form for each exercise. This wiU ensure
maximum benefit and minimize the possibiUty of muscle strain.
Themost productive way to performan anaerobic exercise is to tire
a particular group of muscles as quickly as possible and keep them
tiredduringthe course of theexercise. Thismaysounda Uttle strange,
given that we're aU accustomed totheidea that some athletes workin
precisely the opposite manner —warming up slowly andbuilding to
a fast finish. That maybe fine for a sprint, but we're not talking about
racing here, we're talking about buUding muscle mass. By placing
maximumdemands onyour muscles at first, youput yourselfinto the
anaerobic(or oxygen-deprived) stateright off.Then bygraduaUy pro
gressing to Ughter weights, youforce your muscles to workcontinu
ouslyin the anaerobic stateand thereby buUd them.Thisiswhat I caU
the inverted pyramidapproach to weight Ufting, and for the purpose
of buUding muscle mass, it'sfar superiorto the oldsystem.
Many weight lifters foUow a regimen that requires 10 repetitions
("reps") of a Uft, foUowed bya rest, another 10reps, another rest, and
another 10reps. The rest between each set of repsaUows the heart to
slow, replenishes oxygen to the muscles, andthereby defeats our cen
tral goals. AnaerobicaUy, you mustcontinuaUy keep yourmuscles de
prived of oxygen and force themto develop newmetabolic pathways
that demandless oxygen. The ideais quaUty, not quantity, and it's my
beUef that youcanaccomplish a morethorough andsensible workout
in 30minutesthan youcanin an hour and a halfof conventional, less
strenuous aerobic activity.
I usethe invertedpyramidsystem, socaUed because I start out with
as much resistance as I can handle, and then ease up.
This is how it works: Let's sayyou're performing curls. These in
volve sittingat the edge of a benchor chair and flexing your arms at
the elbows with weights in each hand. You start with the heaviest
Using Exercise to Enhance Insulin Sensitivity 229
weights that you canlift 3-4 times without losing good form. By the
time you've lifted them, say, 4 times, your muscles are tired and you
can't lift them anymore. Youimmediately lift the next Ughter weights
anddo as many reps as you can (say, 3-4), and so on down through
Ughter andUghter weights until youget to atotal of about20reps. You
might find that youcan getout 21, ormaybe youcan onlymanage 19.
That's fine. The idea isthat after the first few repetitions, yourmuscles
are tired and they're workingwhUe they'retired, which is what stimu
lates muscles to buUd more mass.
Exceeding 20reps by much maybe unwise. This isbecause anaero
bicexercise does damage to muscle fibers, and wewant such damage
to be minimal sothat fuU repair wiU have occurred within2 days, or
before youexercise that muscle group again.
Once you've done your reps for a particular muscle group, you
don't needto do that exercise again untU the dayafter tomorrow. You
immediately go on to the next exercise. In this way, you can accom
plish considerablymore in a shorter time frame.
The same system applies to anexercise like sit-ups. Whether you're
doing sit-ups withyour legs straight, bent, orwithoneof those sit-up
boards, youstart off withyourhands behindyourhead. If youcan't do
asingle sit-up with hands behind your head, try it withhands at your
sides or even pressing them on the floor. With practice you'U eventu-
aUy be strong enough to put your hands behind yourhead. If you're
reaUy experienced at sit-upsandhave strong abdominal muscles, you
holdaplate — aflat weight — behind your head. You doas manyrep
etitions as youcan. Maybe you'Uonlyget5-6, maybe only 2-3. Imme
diately put down your weight if you're holding one, ortake your hands
from behindyourheadandfold themacross yourchest. Nowstartdo
ingthe same sit-ups in this fashion. You do as manymorereps as you
can, then put your hands at your sides and do as many more addi
tional repetitions as you can. If you get very experienced and find
yourself doing40ormore sit-ups, it can get prettyboringandis also a
waste of time. When you find yourself doing dozens of sit-ups, you
can get an inclined board or a Roman chair, which is like a sit-up
board but is raised about four feet off theground and permits youto
begin with your head belowyour waist. Again, you foUow the same
tactic. You can also get an abdominal crunch machine with variable
resistance (not those with removable weights that take time to
change). They're the best, but they're expensive. Again, you'd start at
high resistance andwork down for atotal of about 20reps. It is hu-
230 Treatment
manly possible to do more than 4,000 sit-ups at one session, but
what's the point unless you're trying to impress someone or set a
record? Sitting up over a 7-second period and coming down slowly
over 15 or more seconds is actuaUy more effective for exercisingyour
muscles than doing 3 or 4 or more sit-ups in the same time period,
and wiU significantly reduce the total number of sit-ups you can do
before muscle exhaustion.
Cardiovascular exercise is widelyassociated in the public mind with
what the popular press calls aerobicexercise. However, aerobic exer
ciseas many people practice it —a leisurely jog, a relaxing bike ride,
mUd caUsthenics, evena briskwalk—is reaUy of only limited benefit
to your cardiovascular system, doesn't buUd muscles, and has rela
tively little impact on yourstamina andcapacity. Thekindof cardio
vascular exercise I recommendto my patients (and foUow myself) is
verystrenuous, operates intermittentlyin the anaerobic range, and ac-
compUshes tremendous things. Forexample, manyyears ago, before I
becamea physician, I usedto goto diabetes conventions. Therewasal
ways a groupof doctors whowould get up in the morning, don their
running togs, and go running. These were peoplewho ran everyday.
I'm not a runner; I work out in the gymeveryday. But I do a particu
lar cardiovascular workout on a recumbent exercise bicyclethat I wiU
explain. I wouldgo out with these doctors on their runs. After a few
mUes, peoplewouldstart droppingout. EventuaUy, I'd be the onlyone
left — and then I'd go another five mUes and come back. Clearly, al
though I was older than most of these people, and not a runner, I had
much more stamina. The stamina was created by this anaerobic car
diovascular exercise.
Exercise Harder, Exercise Better
Cardiovascular workouts can be performed on a treadmiU, a stair
climber, or bicycle. If you're female, I'd recommend a treadmiU, be
causerunning impactsyour feet and thus helpsincrease bone density
in your legs. However, if done to excess or with inadequate arch sup
ports, the impact can injureyour knees. If you'remale, I recommend
a recumbent bicycle rather than the standard upright bike; it's much
more comfortable for men because the seat is luce an ordinary chair.
Using Exercise toEnhance Insulin Sensitivity 231
IdeaUy, your machine should haveameter that reads the amount of
work that you're doingin calories (or joules) perminute asweU asto
tal calories (or joules), but certainly youcan get agood workoutwith
just amUeage meter. It is important to wear a pulse meter.The brand
thatI like bestiscaUed Polar; it costs about $60, and youwear asensor
around your chest withawristwatch-type readout. If youbelong to a
health clubthat hasatreadmiU with a pulse meter in the handlebars,
youwon'thave to put oneonyour chest, but some sortof pulse meter
is essential. The degree of workout you're getting is measured by how
fast your heart works. If youget evaluated byacardiologist before you
start your exercise program, you shouldaskhim or her what your ini
tial target pulse rate ought to be. Over time, youcanincrease it.
There's a formula that weuseto specify atheoretical maximum at
tainable pulse rate: we take 220 and subtract from it your age. So if
you're sixtyyears old, you'd have atheoretical maximumpulse rate of
160 — that is, in theory, you shouldn't be able to exercise at a faster
pulse rate. Your doctor wiU decide based on your overaU healthand
fitness level what percentage of this would beagood initial target rate
for you— say, 75-80 percent of maximum. Rarely would a doctor
start you out at 85 percent of maximum or higherif you werenot in
shape. EventuaUy, youmay find that you can get up to and beyond
your theoretical maximum— I can exercise at 160 even though my
theoretical maximumis 148.1 can do this without having aheart at
tack in part because I've been exercising strenuously for forty years.
Don't expect — even after years of this kindof exercise — to getyour
heart rate up to or even near your theoretical maximum, or to your
target, right after youbegin thiskindof workout. It takes time. I get to
my target pulse rate at the endof about 10 minutesof trying.
To do a reaUy effective anaerobic/cardiovascular workout, you use
thesame principles as youuse lifting weights. Start out by selecting a
safe, comfortable speed andsetting the resistance of yourmachine to
the point where your muscles are so tired after about 20 seconds that
youcan'tgoany further. As soonas you reach this point, lower the re
sistance setting sUghtly and keep going. For treadmiUs, the resistance
wiU be the angle at which you're running uphiU. So if you're using a
treadmiU, youneedto be able to setthe incline of yourtreadmiU from
the handlebars — youdon't want to get off, reset the angle, thenget
back on. You'U lose yourrhythm, regain someof the oxygen in your
musclesand heart, and defeat the point of the workout.
As with the weights, you lower the resistance a Uttle at a time, and
232 Treatment
eachtime you lowerit, shoot for 20seconds of exercise untUyou can't
goanymore. Nearlyfromthebeginningyou're wipedout, yetyoukeep
doingit at lower and lower resistance. This is a realworkout.
Your goal wiU be to get your heart rate up to (but not above) the
training level recommended byyour physician. If you can't reach the
recommended rate whenyoucanbarely noticethe resistance of your
machine, increase your speeduntil you get to your pulsetarget. Tryto
maintain this rate for up to 5 minutes.
A major goal of cardiovascular exercise is to enhance your heart
rate recovery time, i.e., to shorten it. (Cardiologists now beUeve that
the faster your heart rate slows fromyour target to near resting rate,
the better your cardiacfitness.) Aminimal test of recoverywouldbe to
slowyour heart rate by 42 beats per minute from your maximum
within 2 minutes of slowing your feet until you are barelywalkingor
pedaling. Todo this you sprint at top speedfor 2-5 minutes and then
drop to a veryslowspeedfor 3-5 minutes,and then sprint again, over
and over.
When you think you've had enough, lowerthe resistance to zerobut
keep your legs moving veryslowly until your pulsehas returned to a
value about 30 percent above your starting point. This slowexercise
pumps blood back to your heart fromyour legs, thereby greatly re
ducing the hazard of a postworkoutheart attack.
I recommend that rather than timing your workout, you look at the
calorie counter on the machine, if it has one, and decide on a particu
lar number of calories that you want to shoot for. Calories are a mea
sure of work done and therefore a reasonable gauge of your workout.
Minutes or even mUes don't take effort into account. I aim for about
200calories. When it getsup to that range, I caU it quits. But the point
of this kind of exercise isn't weight loss, so don't start looking at the
caloriecounter thinking that if you burn 200more caloriesyou'U lose
another pound — exercise just doesn't work that way. IncidentaUy, I
have a retired patient who actuaUy has the time and the stamina to
continue intermittent sprinting for an hour.
If you're doing cardiovascular exercise of this type, you haveto be very
careful, especiaUy if you're a long-term diabetic, or a recent-onset dia
betic over the ageof forty,or havea famUy history of coronary disease.
Using Exercise to Enhance Insulin Sensitivity 233
One rule isthatyouneverfinish acardiovascular workout and stop cold.
I had an overweight nondiabetic cousin who started jogging when he
was about fifty years old. He was in his second month of exercising,
not doing anything more than jogging with friends. One day, after
they stoppedjogging, he dropped dead of a heart attack. He and his
jogging buddies were in the habit of stopping cold after theirrun to
chat. Stopping cold isan extremely bad idea — if people are going to
drop dead of heart attacks from running orbiking, it's most often im
mediately after the exercise thatthis happens. Why?
WhUe you're exercising, your heart isbeatingveryrapidly because it
and your legs require alot of blood. By pumping your legs up and
down, you're pumping blood from your legs back to your heart. The
muscles thatare demanding alot of blood are bothin yourlegs andin
your heart, but theblood's getting pumped back toyour heart byrun
ning. If youstop cold, your muscles are stiU going to demand alot of
blood — they've been depleted of oxygen and glucose — and gravity
isgoing tohelp themget theblood. The problem is, they're nolonger
pumping theblood back totheheart. Suddenly your heart isdeprived
and, if your coronary arteries are narrowed byatherosclerosis, you're
set up for aheart attack.
Whether you're on a treadmiU, bike, or stair climber, cut the resis
tance setting to zero andproceed at averyslow pace after yourwork
out until your heart rate slowly comes down to no higher than about
30 percent above your initial starting rate. If your resting pulse is 78,
you don't want to stop your biking, walking, or step cUmbing untU
your heart rate is 101 or below.
As your strength and endurance increase for anyexercise, it wiU be
comeprogressively easier to perform. If it becomes too easy, youwon't
get any stronger. The key to getting progressively morestrength and
endurance is to make the exercise progressively more difficult. This
canbe done for almostanyactivity.
If youare Ufting weights, for example, every few weeks (ormonths)
you can addaverysmaU weight (say a separate 2^-pound plate) to
the weight stack for anyexercise. When doing acardiovascular exer
cise, you might try to increase your maximum heart rate by, say, 2
beats per minute every 2months, byincreasing your resistance setting
234 Treatment
or your speed. A swimmer canassign a fixed time period, say30 min
utes, for doinglaps. The goal wouldbe to graduaUy increase the num
ber of laps. Thus, afteramonth you might increase your speedto get
15V2 lapsinsteadof 15lapsin 30minutes, and so on. Of course, awa
terproof wristwatch would be helpful.
Evenwalkingcan evolveinto both an enduranceand a bodybuUd
ing activity. AU you needis awristwatch, a few Ughtweight dumbbeUs,
and a pedometer. The pedometer is a small gadget from a sporting
goods store that you cUp onto your belt. It measures distance by
countingyour steps. Suppose youwishto set aside 30minutes per ses
sion for walking. You begin by walkingat aleisurelypace for 15min
utes and then returning at the same pace. Recordyour distance from
the pedometer.Thereafter, try to walkat least that distance in the same
time period. After ten sessions, you might try to increase distanceby 5
percent over the same time. If you increase distance by this amount
every ten sessions, you'U eventuaUy find yourself running. You can
then graduaUy increase your running speedin the same fashion.
Supposeyour doctorhastold you not to run because of abadknee
or fragUe retinal blood vessels. Limit your speed to a fast walk, but
start swinging your arms a Uttle bit. Over time, try swinging them
higher and higher. When you think they are going so high that you
look siUy, start with the dumbbeUs. You might begin with a pair of
1-pound dumbbeUs and short swingsof the arms. Wear gloves if the
dumbbeUs feel cold. Again,graduaUy increase the distance you swing.
When you eventuaUy feel youlook siUy, try 2-pound dumbbeUs. After
a yearor two, you may be goingat avery fast walk, swinging 5-pound
(or even heavier) dumbbeUs. Imagine what your physique wiUlook
like then. You'U alsoprobablyfeel younger and healthier.
The exercises I've mentioned above are by no means the only ones.
There are countless different ways you can exercise — voUeybaU,
snowboarding, surf-kayaking, cross-country skiing, you name it. The
most important considerations are keeping within the restrictions
your physician might place on your activity, and discovering what you
like to do best — and stickingwith it. After that, aU you have to do is
monitor and correct your blood sugars, record the exercise on your
Glucograf form, and keep exercising in a progressive fashion. The
payoff—longer life, lowerstress, weight lossif you'reoverweight, and
better overaUhealth — is usuaUy worth the time and effort.
Oral InsuHn*Sensitizing Agents,
Insulin~Mimetic Agents,
and Other Options
Ifdietandexercise are not adequate tobring your bloodsugars un
der control, the next level of treatment to consider is oral blood
sugar-lowering medication, commonly known as oral hypo
glycemicagents (OHAs).
There are three categories of OHAs, those that increase sensitivity
to insulin, those whose action resembles that of insuUn, and those that
provoke your pancreas to produce more insulin. The first group is
known as insulin sensitizers (or ISAs, for insulin-sensitizing agents);
the second are the insulin mimetics (or IMAs, for insulin-mimetic
agents), which act like insulinbut do not buUd fat. FinaUy, there are
the original OHAs, like sulfonylureas.
I only recommend the insuUn sensitizers and insulin mimetics, for
reasons that wiU become plain in shortorder. (Some drug companies
have combined pancreas-provoking OHAs with insulin sensitizers, a
moveI strongly chaUenge. TeU yourdoctor youdo not want anyprod
uct containingan agent that works by causing the pancreas to make
more insulin.This includes the old sulfonylureas andthe new, simUar
drugs caUed megUtmides and phenylalanine derivatives/)
For people who stUl have sufficient insuUn-producing capacity, in
sulinsensitizers alone may provide the extra help they need to reach
their blood sugar target. Some insulin-resistant individuals who pro-
*In addition to causing betacell burnout, sulfonylureas also impair circulation
in the heart and elsewhere by closing ATP-sensitive potassiumchannelsthat re
laxbloodvessels. Theyhave beenshown to increase aU causes of mortaUty, in
cluding deaths from heart diseaseand cancer.
236 Treatment
duce Uttle or no insuUn on their own may find a combination of in
sulin sensitizersand insulin mimetics useful in reducing their doses of
injected insulin.
There are three ISAs currentlyon the market, and at this writing I
prescribeaU three of them— metformin (Glucophage), rosigUtazone
(Avandia), and piogUtazone (Actos). RosigUtazone and piogUtazone
have simUar effects upon blood sugar, so it servesno purpose for one
individual to use both.
Anote: Sincebrand names vary from country to country, I wiU use
only the generic names in my discussion of drugs in this chapter. In
my experience, however, not all genericmetformins match the effec
tiveness ofGlucophage.
Some of the OHAs on the market are not insulin-sensitizing or
-mimetic. Instead, theyprovokethe pancreasto produce more insulin.
For several reasons, this is considerably less desirable than taking a
medicationthat sensitizes youto insulin.First,the pancreas-provoking
OHAscan causedangerously lowblood sugar levels (hypoglycemia) if
used improperly or if meals are skipped or delayed. Furthermore,
forcing an already overworked pancreas to produceyet more insulin
can lead to the burnout of remainingbeta ceUs. These products also
facUitate beta ceU destruction by increasinglevelsof a toxic substance
caUed amyloid. FinaUy, it has been repeatedlyshown in experiments —
and I have seen it in my own patients — that controlling diabetes
through blood sugar normaUzation can help restore weakened or
damaged beta ceUs. It makes absolutely no senseto prescribe or rec
ommend agents that wiU cause them renewed damage. In a nutsheU,
pancreas-provoking drugs are counterproductive and no longer have
anyplacein the sensible treatment of diabetes.
As it's far more productive to talk about good medicine, I wiU leave
pancreas-provoking OHAsin the past, whereeventhe newer ones be
long, and from here on out discussonly insuUn sensitizersand insulin
mimetics. Then, at the end of the chapter, I wiU look at possible new
treatment options for three specialcircumstances.
The great advantage of insulin sensitizers is that they help to reduce
blood sugar by making the body's tissues more sensitive to insulin,
OralInsulin-SensitizingAgents, Insulin-Mimetic Agents, andOtherOptions 237
whether it's the body's own or injected. This is a benefit that can't
be underestimated. Not only is it a boon to those trying to get their
blood sugars under control, but it's also quite useful to those who are
obese and simultaneously trying to get their weight down. By helping
to reduce the amount of extra insulin in the bloodstream at any given
time, these drugs can help aUeviate the powerful fat-buUding proper
ties of insulin. I have patients who arenot diabeticbut have come to
me for treatment of their obesity. Insulin sensitizers have been a real
plus to the weight-lossefforts of some because of their abiUty to cur-
taU insulin resistance. Their major shortcoming is that they're rather
slow to act — for example, they wiU not prevent a blood sugar rise
from a meal if taken an hour before eating, as some of the beta
ceU-pushing medications wiU. As you wiU learn, however, this can be
Some obese diabetic patients come to me who are injecting very
large dosesofinsuUnbecausetheir obesity makes themhighly insulin-
resistant. These high doses of insulin faciUtate fat storage, and weight
loss becomes more difficult. Insulin sensitizers make these patients
more sensitive to the insulin they're injecting. In a typical case I had a
patient taking 27 units of insulin at bedtime, even though he was on
our low-carbohydrate diet. After he startedon metformin, he was able
to cut the dose to about 20 units. This is still avery high dose, but the
metformin faciUtated the reduction.
Insulin sensitizers have also been shown to improve a number of
measurable cardiac risk factors, including blood clotting tendency,
lipid profile,Upoprotein(a), serum fibrinogen, blood pressure,C-reac
tive protein, and even abnormal thickening of the heart muscle. In ad
dition, metformin hasbeen found to inhibit the destructivebinding of
glucose to proteins throughout the body— independent of its effect
upon blood sugar. It has been shown to reduceabsorption of dietary
glucose, and also improves circulation, reduces oxidative stress, re
duces blood vessel leakage — in the eyes and kidneys — and reduces
the growth of fragUe new blood vessels in the eyes. It has also been
shown to improve satiety in women near menopause. Thiazolidine-
diones such asrosigUtazone and piogUtazone canslowthe progression
of diabetic kidney disease, independent of their effects on blood sug
ars. These medications can also down-regulatethe genesthat cause fat
storage, and they havebeen found to delayor prevent the onset of di
abetesin some high-risk individuals.
238 Treatment
In addition to the insulin sensitizers, there are some substances sold in
the United States as dietary supplements that are effective for helping
to control blood sugars. Manystudies in Germany have demonstrated
this effectfromR-alphalipoicacid,or ALA. A2001 study showedit to
work in muscle and fat ceUs by mobilizing and activating glucose
transporters — in other words, it works like insulin, or is an insulin
mimetic. German studies have also shown that its effectiveness in
mimicking the effects of insulin is greatlyenhanced when used with
equivalent amounts of eveningprimrose oU, another dietary supple
ment. ALA can reduce the body's natural levels of biotin, so it should
be taken in a preparation that contains biotin (see footnote below).
ALA and eveningprimrose oU are no substitute, however, for injected
insulin — they are at best a fraction as potent. Still,their combined ef
fectiveness is significant.
AdditionaUy, ALA is perhaps the most potent antioxidant on the
market and has certain cardiovascular benefits simUar to those
claimed for fish oU. Many of the cardiologistswho were taking vita
min Efor its antioxidant propertiesten yearsagoare nowtakingALA.
I've been taking it myself for about eight years. When I began, I
promptly found that I had to lower my insuUn doses by about one-
third. R-ALA and evening primrose oU do not appear to mimic one
important property of insulin—they don't appear to facUitate fat
storage. They are both avaUable without prescription from some
health food stores and from some pharmacies.* They have the po
tential to cause hypoglycemia in diabetics who inject insulin if they
don't adjust their insulindosages accordingly. I havenever seen them
cause hypoglycemia, however, when they are not used with injected
Other German studies have shown dramatic improvements in
diabetic neuropathy (nerve damage) when alpha Upoic acid is ad-
*Although conventionalALA is widely available, R-ALA is more effective. As of
this writing, the principal manufacturer in the United States isGlucoreU Inc.,of
Orlando, Florida, phone (866) 467-8569, www.insulow.com. Their product In-
sulowcontains 100mgR-ALA per capsule, plus 750meg(0.75mg) biotin. Insu-
low is also available from RosedalePharmacy.
OralInsulin-SensitizingAgents, Insulin-MimeticAgents, andOtherOptions 239
ministered intravenously in large doses over several weeks. Given its
antioxidant and likely anti-inflammatory properties, this isn't that
surprising. But it faUs under the category of "Don't Try This at
AlphaUpoic acid,likehigh-dose vitaminE(the formcaUed gamma
tocopherol) and metformin, can impede glycosylation and glycation
of proteins, both of which cause many diabetic compUcations when
blood sugars are elevated.
I usuaUy recommend two 100mg tablets every 8 hours or so, with
one 500 mg capsule of evening primrose oU at the same time. If an
insulin-resistant patient is alreadytaking insulin, I wUl start her on
half this doseoncedaUy and observe bloodglucose profiles and lower
insulin dose as I raise alpha Upoic acid and evening primrose oU.
Again, it's aU trial and error.
GeneraUy speaking, these agents are natural choices for a type 2
diabetic who despite a low-carbohydrate diet cannot get his weight
down or his blood sugars into normal ranges. The blood sugar ele
vation may be limited to a particular time of the day, it may be dur
ing the night, or it may entaU a sUght elevation aU day. We base our
prescription on the individual's blood sugar profiles. If even on our
diet, blood sugar exceeds 300 mg/dl at any time of the day, I'll im
mediately prescribe insulin and won't even attempt to use these
agents, except to eventuaUy reduce doses of injected insulin. If your
blood sugar is higher upon arising than at bedtime,we'dgive you the
sustained-release version of metformin at bedtime. If your blood
sugar goes up after a particular meal, we'dgive you a relatively rapid
acting insulin sensitizer (rosigUtazone) about 2 hours before that
meal. Since food enhances the absorption of the thiazoUdinediones,
wemight give it with the meal. If bloodsugars are sUghtiy elevated aU
daylong,wemight use alphaUpoic acidand evening primroseoU on
arising, postiunch, and postdinner. It should be noted, however, that
the ISAs are considerably more effective than IMAs at loweringblood
240 Treatment
Let'ssayyou're a type 2 diabeticand through weight loss, exercise, and
diet, you pretty much have your blood sugars within your target
range. StiU, your blood sugar profiles showa regular elevation in the
mornings after a low-carbohydrate breakfast, probably due to the
dawn phenomenon.
Of the medications I've describedhere, the most rapid to start act
ing is rosigUtazone, which, although it reaches peak levels in the
bloodstream in about an hour, probably achieves its fuU effect after
about 2 hours. Soyou might takea starting dose of 4 mg upon arising
and then eat breakfast 1-2 hours later. If this is only partly effective,
the dose can be increased to two 4 mg tablets or one 8 mg tablet (the
maximum recommended daUy dose). If this is somewhat effective, but
2 hours after breakfast your blood sugars are still above target, you
might add an extended-release dose of metformin before you go to
bed. This type of metformin achieves its peak blood levels after about
7 hours. A starting point would be one 500 mg tablet at bedtime. If
this stiU doesn't get your blood sugars into target range, then you
could increase the dose graduaUy, perhaps by one more tablet at bed
time for a week and so on, until you reach a maximum of 4 tablets a
night or you hit your target. I always recommend the least possible
dosage— partly due to the Laws of SmaU Numbers, but also because
of the reduction of likelihood for potential side effects. With met
formin, if you buUdup your dosageslowly, it lessensthe possibUity of
gastrointestinal discomfort that about one-third of users of the older,
more-rapid-acting version experience.
In some cases, blood sugar levels either increase overnight or in
crease during the first 2 hours after you arise. The latter situation is
most likely due to the dawn phenomenon. Either situation may re
spond to timed-release versions of metformin (Glucophage XRin the
United States) with or without ALA plus evening primrose oU, aU
taken at bedtime, using the doses described above. If need be, piogU
tazone may also be added at bedtime. Tablets of piogUtazoneare sold
in 15mg and 45 mg doses.The maximum daUy dose is 45 mg.
Another possibUity that wouldwarrant oral medicationwould be if
your blood sugar levels increasedafter lunch or dinner. Wecould pos
siblycover the problem meal with rosigUtazone by taking it 1-2 hours
before eating.
OralInsulin-SensitizingAgents, Insulin-MimeticAgents, andOtherOptions 241
TABLE 15-1
U.S. brand Available (effective)
name dosages dally dosage
Metformin InsuUn Glucophage 500,850, 2,500mgf
sensitizer and generic 1,000 mg*
Metformin InsuUn Glucophage XR 500 mg 2,000mg1
extended sensitizer and generic
RosigUtazone Insulin
Avandia 4,8mg 8mg
PiogUtazone InsuUn
Actos 15,30,45 mg 45 mg
R-alphaUpoic InsuUn Insulow 100 mg 1,800 mg
acid (ALA) mimetic
with biotin
Evening InsuUn Many 500 mg 3,000 mg
primrose oil mimetic recommended
(EPO) booster
for every
300 mg
of ALA
*Alsoavailableas a Uquid.
f Forreasons not apparent to me,the manufacturer's recommendation for maxi
mum dailydosing is lessfor extended-release metformin than for the standard ver
Sulfonylurea and the newer gUtazar OHAs carry the veryreal possi
bUity of causing dangerously lowblood sugars, which is one of the
reasons I never prescribe them. However, this is only remotely likely
with the insulin-sensitizing and insulin-mimetic agentsUsted above.
Noneof theminterferes withthe self-regulating system of a pancreas
that can stiU makeits own insulin. If your blood sugar drops too low,
242 Treatment
your body wiU most likelyjust stop makinginsulin automaticaUy. Sul
fonylureas and simUar drugs, on the other hand, because they stimu
late insulin production whether the body needs it or not, can cause
Although the manufacturer and the scientific literature claim that
metformin does not causehypoglycemia, I did havea patient who ex
periencedhypoglycemia. Shewasveryobese but onlyverymUdly dia
betic, and I was givingher metformin to reduce insulin resistance to
facUitate weight loss. When I put her on metformin, her blood sugars
went too low (but not dangerously) — down into the 60s.
So there may be some very sUght risk of hypoglycemia with the
insulin sensitizers or insulin mimetics, but this is not at aU compara
ble to the great risk with the sulfonylureas and simUar medications.
One warning, however. The bodycannot turn off injectedinsulin, so
if you are taking insulin plus anyof theseoral agents,hypoglycemia is
If theseagentsare not adequateto normalizeblood sugarscompletely,
chances are there is something awry in the diet or exercise portion of
your treatment program. The most likely culprit for continued ele
vatedblood sugarsis that the carbohydrate portion of your diet is not
properly controUed. So the first step is to examine your diet again to
seeif that's wherethe problemUes. With manypatients, this is a mat
ter of carbohydrate craving. If this is the case and your carbohydrate
craving is overwhelming, I'd recommend that you rereadChapter 13
and consider pursuing one of the techniquesdescribedthere. If diet is
not the culprit, then the next thing — no matter how obese or resis
tant to exercise you might be —wouldbe to try to get you started on
a strenuous exercise program. If even this doesn't do the trick, we'U
It's also worth keeping in mind that infection or illness can seri
ouslyimpair your efforts at bloodsugarnormalization. If your blood
sugarlevels areway out of lineeven withthe useof insulin, youmight
alsoconsider talkingto your physician about potential underlying in
fection, especiaUy in the mouth (see pages 100-101).
OralInsulin-SensitizingAgents, Insulin-MimeticAgents, andOtherOptions 243
AlthoughinsuUn mimeticsand insuUn-sensitizing agents are some of
the best tools we havefor controUing blood sugars, they are not with
out their difficulties. Sincealpha Upoic acid and evening primrose oU
are not prescription drugs in most countries (Germany is a notable
exception), they are not covered by most health insurance. Alpha
Upoic acid is not inexpensive; at this writing, a supplyof 180Insulow
100mg tablets costs about $30.
ALA reduces body stores of biotin, a substance that aids in the uti
lization of protein and a varietyof other nutrients, so when you take
alpha Upoic acid, you might be wiseto takebiotin supplements also—
unless you are taking Insulow,which alreadycontains biotin. Yourbi
otin intake should theoreticaUy equal about 1 percent of your alpha
Upoic acidintake,soif youaretaking1,800 mgALAper day, in theory
youwouldtakeabout 18mgof biotin. Mostof mypatientswhouseal
pha Upoic acid don't take more than about 15mg biotin per day, and
they experienceno apparent adverse effects. Most preparations come
only in 1mg strengths.
Metformin has a very low side-effects profile, with the exception
of gastrointestinal distress —queasiness, nausea, diarrhea, or a slight
beUyache — in as many as a third of the people who try the non-
extended-release version. Most people who experience such discom
fort, however, find that it diminishes as they become accustomed to
the medication. Only a very fewpatients can't tolerate it at aU. (Some
patients, particularly obese people who are anxious to achieve the
weight loss that metformin can facUitate, wiU ignore any initial gas
trointestinal distress and use an antacid drug such as Pepcid or Taga
met for reUef. Others, who may only experience relatively mUd
discomfort, are wiUing to tolerate it for a few weeks just to get things
roUing.) Rarecasesof diarrhea have been reported long after the start
of metformin therapy. Theywere reversed by discontinuation of the
medication. I havenot observedgastrointestinal sideeffects associated
with the use of thiazolidinediones or extended-release metformin.
Metformin's predecessor, phenformin, was, in the 1950s, associated
with a potentiaUylife-threateningcondition calledlacticacidosis. This
occurred in a smaU number of patients who were already suffering
fromheart faUure or advanced Uver or kidney disease. AlthoughI have
244 Treatment
read of only a few instances of lactic acidosis associated with met
formin, the FDAadvises against using it in individualswith these con
ditions. Metformin has been reportedto lowervitamin B-12 stores in
about one-third of users. This effect canbe preventedby taking a cal
cium supplement (see page 179).*
The two thiazoUdinediones currently avaUable in the United States
both have potential for minor problems. PiogUtazone is clearedfrom
the bloodstream by the liver, utilizing the same enzyme it utilizes to
clear many other common medications. The competition for this en
zyme can leave dangerously elevated blood levels of some of these
drugs. If you are taking one or more of these competing medications,
such as some antidepressants, antifungal agents, certain antibiotics,
and others, you should likely not be using piogUtazone. You should
check the package insert for potential drug interactions and talk to
your physicianand pharmacist.
RosigUtazone and especiaUy piogUtazone cancause a smaU amount
of fluid retention in some people.The consequenceof this is adilution
of redblood ceU count andmUd swellingin the legs. I'veseenanumber
of such cases. There can alsobe a smaUweight gaindue to the retained
water, not to fat. This water retention has been associated with a few in
stances of heart faUure in individuals taking one of these medications
plus insulin. In the UnitedStates, the FDA hastherefore recommended
that doses of these agents not exceed 4 mg and 30 mg per day, respec
tively, for peoplewho inject insulin. I havetreated many insulin users
with them and haveseensUght swelling ofthe legsin some cases. When
this occurred, I discontinued the medication immediately. There also
have been very rare cases of reversible Uver damage associated with
both rosigUtazone and piogUtazone.* A study reported in Endocrine
Practice in 2001 showedasignificant increase in serumtriglyceride lev
els for users of rosigUtazone but not piogUtazone. On the other hand,
piogUtazone hasbeenshownto improveUpid profiles (LDL, HDL, and
triglycerides), whUe rosigUtazone cancause asUght impairment.
* A deficit ofvitamin B-12 can increase serum levels of the renal disease risk fac
tor homocysteine. It would therefore be wise for your physician to check your
serum homocysteine everysixmonths whileyou are using metformin.
*Eventhough reports of Uver toxicityare far fewer than with some commonly
used medications such as niacin and the so-caUed statins, it's a good idea for
users of these insulin sensitizers to have their blood tested for liver enzymes
Oral Insulin-SensitizingAgents, Insulin-Mimetic Agents, andOther Options 245
Because of the possibUity of fluid retention, neither medication
should be used by patients with significant cardiac, lung, or kidney
disease, or with any degreeof heart failure.
I usuaUy start people on rosigUtazone to avoid potential competi
tion for clearance by the liver with other drugs another physician
might prescribe in the future.
Metformin works principaUy by lowering insuUn resistance in the
liver. It also impairs, somewhat, theabsorption of carbohydrate bythe
intestine. Thiazolidinediones principaUy affect muscleand fat, and to
a lesser degree the liver. Thus, if metformin does not fuUy normalize
blood sugars, it makes senseto add one of the thiazoUdinediones —
andvice versa. Since rosigUtazone andpiogUtazone workbythe same
mechanisms, it makes little sense to use both in the same individual.
The FDA suggests that doses of piogUtazone not exceed 30 mg daUy
when taken with metformin.
Since ALA and eveningprimrose oU work as insulin mimetics, it is
certainlyappropriate to add these to any combination of the other
Thethiazolidinediones donot have theirfuU bloodsugar-lowering ef
fects on thedaytheyarestarted. PiogUtazone achieves its fuU potency
after a few weeks, and rosigUtazone mayrequire up to twelve weeks.
When bloodsugars aremuch higher thanthe targets that I set,both
metformin and the thiazoUdinediones cancause the pancreas to in
crease its insulin production in response to glucose. Because of the
lower bloodsugars that wesee, thiseffect becomes insignificant.
Vitamin Asupplementation has been shown to lower insulin resis
tance (as does vitamin E)* in doses of about 25,000 IU daUy. Since
slightly higher doses of vitamin AarepotentiaUy very toxic, and doses
as low as 5,000 IU can cause calcium loss from bone, I would consider
*Vitamin Eshouldonlybe usedintheforms called gamma tocopherol or mixed
246 Treatment
only moderate doses of its nontoxic precursor, beta carotene, for this
Studies have shown that magnesium deficiency can cause insuUn
resistance. It would therefore be a sensible idea for physicians to test
type 2 diabetics for red blood ceU magnesium (not serum magne
sium) levels. If the level is low, magnesium supplementation should
help. I recommend a product caUed slow-mag in smaUdosesthat can
be increased if the test remains low after one month. Excessive doses
cancausediarrhea. Sinceredblood ceU magnesium is not a perfect in
dicator of blood magnesium stores, and since magnesium supple
ments are benignto peoplewith normalkidneys (except fordiarrhea),
it is appropriate to use magnesium supplements as a test to see if
blood sugars decUne. Doses as high as700 mg daUy are common for
SimUarly, zinc deficiency can cause diminished production of lep
tin, a hormone that impedes overeating and weight gain. Such defi
ciencycanalso impair functioning of the thyroid gland. It is thus wise
for aU type 2 patients to ask their physicians to test their serum zinc
levels and to prescribe zinc supplementation if warranted. FoUow-up
serum zinc levels should be measured to ensure that normal levels are
not exceeded.
Compounds of the heavymetalvanadiumhavebeenshown to lower
insulin resistance, reduce appetite, and possibly also act as insuUn-
mimetic agents. They arequite potent in lowering blood sugars, but
there's a catch.Vanadium compounds work by inhibiting the enzyme
tyrosine phosphatase, which is essential to many vital biochemical
processes in the body. The possibUity is quite real that this inhibi
tion can be damaging. Since clinical trials in humans have not ex
ceededthree weeks in duration, long-term freedom from adverse ef
fects has yet to be documented. Some users of vanadium compounds
have experienced gastrointestinal irritation. Although vanadyl sulfate
is widely avaUable in health food stores as a dietary supplement and
hasbeen used for years without anyreports of adverse effectsin med
ical journals, I tentatively recommendthat it be avoided until more is
Except bycommercial pilotswhomust avoidinsuUn (seepage247).
OralInsulin-SensitizingAgents, Insulin-MimeticAgents, andOtherOptions 247
In theory at least, there are individuals who do not respond to any of
the measures recommended in Chapter 13for the control of carbohy
drate cravingand overeating. Thesepeoplecan be helpedslightly by a
product caUed acarbose (Precose). Acarbose is avaUable as 25 mg, 50
mg, and 100 mg tablets to inhibit action of enzymes that digest
starches and table sugar,* thereby slowing or reducing the effects of
these no-no foods upon blood sugars. It is interesting that the ADA
recommends eating starches and sugars and then the simultaneous
use of acarboseto prevent their digestion.
The maximum recommended daUy dose of acarbose is 300mg. It is
usuaUy taken at the time of carbohydrate consumption. Its major ad
verse effect, in about 75percent of users, isflatulence (predictably), so
it is wiseto taper up the dosegraduaUy. It should not be used for pa
tients with any intestinal disorders (e.g., gastroparesis). I have never
had the need to prescribeit.
Commercialairline pUots with diabetes are currently facedwith regu
lations in the United States that threaten loss of Ucense (and UveU-
hood) if they inject insulin. Certainlythese people should try aU of the
oral agents recommended above, as weU as a low-carbohydrate diet
and strenuous exercise. They should also consider vanadyl sulfate,
magnesium, and the other supplements listedunder "Other Consider
ations" on page 245.
There is yet another potentiaUy powerful way of lowering insuUn
resistance for people with this problem. It's been demonstrated that
men whose body iron stores placed them in the top 20 percent of
the nonanemic population had much greater insuUn resistance than
those in the bottom 20 percent. Furthermore, their insulin resistance
*The enzymes are alpha-glucosidase and pancreatic amylase.
248 Treatment
dropped dramaticaUy when they donated enough blood every two
months to keepthem in the bottom 20percent. I've actuaUy seenthis
work on my own patients. Agood measure of total body iron is the
serum ferritin test. Sincesome blood banks wiU not accept blood do
nations from diabetics, it may be necessary to visit a hematologist for
a phlebotomy (removal of blood from a vein) everytwo months. It is
likely that most insurance planswUl paythe hematologist's fee.
Women are much lesslikely than men to havehigh normal ferritin
The recent avaUabUity of the first DPP-4 inhibitor (seepage205) pro
vides one more option for those who refuse to take injections. This
new product, sitagUptin (Januvia) comes in piU form— 25, 50, and
100mg. The maximumadult dosefor peoplewithout kidneyimpair
ment is 100 mg once daUy. It wiU significantly reduce the effect of
glucagon upon blood sugar during and after meals (Chinese restau
rant effect). It can be used as part of a three-way combination with
metformin and a thiazolidinedione.
Insulin: The Basics of SelMnjection
Asyou may have learned from the preceding chapter, certain
oral agents, such as ISAs and insuUn mimetics, are valuable
for controlling blood sugars but can onlygo so far. If you're
taking the maximum effective doses of oral agents and your blood
sugars remain elevated — in spite of diet, exercise (where feasible),
and weight loss — injected insulin wiU be essential to bringing your
bloodsugars downto yourtarget range.*
Although many patients initiaUy balk at the idea of injecting in
sulin, you shouldlook at this as an opportunity, not a curse, because
insulin injections will increase the likelihood that you can bring about a
partial recovery ofyour pancreatic beta cell function. This is especiaUy
trueif youare aslimtype2or arecently diagnosed type 1.
If you're afraid of insuUn because youimagine that onceyoustart,
you'Unever beable to stop, you've faUen victimtoacommon myth.In
reaUty, injected insulin is the best means we have at this writing for
preventing beta ceU burnout.
The Biostator GCIIS, an "artificial pancreas," was a device devel-
*Investigators inBuffalo, NewYork, have demonstrated that injected insuUn ap
pears to lower the production of inflammatory substances and increase levels of
anti-inflammatoryagentsin obeseindividuals. Since inflammationincreases the
UkeUhood of atherosclerosis, chronic useof insuUn injections can lower riskof
cardiac disease, peripheral vascular disease, andstroke, independent of itseffects
uponblood sugar. Injected insulin also facUitates thedilation (opening) of coro
nary and other arteriesthat maybe constricted in manydiabetics and evenin
nondiabetics. It also has been found to improve the absorption of oxygen by
blood flowing through the lungs.
250 Treatment
oped in the 1970s when the average insulin-usingdiabetic took a sin
gle, daily, industrial dose of insulin. The device maystill be available.
In any case, its initials stood for "glucose-controlled insulin infusion
system." That's exactly what it aimed to do — infuseinsulin as a pan
creas would, based on blood glucose levels. It attached to the patient
through two intravenous connections, one that measured blood sug
ars constantly and another that deliveredglucose or insulin to correct
blood sugars to 90 mg/dl virtually instantaneously. Although it was
not practical for home use (it had a staff of two— one to operate the
machine and one to service it — and rented for tens of thousands of
dollars a month), it did useful research, the most important element of
which was that it showed that beta cell burnout could be reversed or
halted, even by relatively short exposure to normalized blood sugars.
Many years ago, Gerald Reaven, MD, author of Syndrome X, con
ducted a study with thirty-two diabetics, half of them female, half of
them male. One at a time, he put them into the hospital and had them
attached to the Biostator for two weeks. His staffchecked HgbA,c on
arrival, at discharge, and every three months thereafter. They found
that HgbA,c plummeted during the two-week treatment period, but
the most important thing they found was that when the subjects went
back to their ordinary lives and their poor diets, their HgbAJC mea
sures took an average of two years to return to their high, pretreat-
ment values.
Considerable beta cell recoveryclearlyoccurred after just two weeks
of normal blood sugars. In fact, it took two years to undo those two
weeks of healing. I'm not inviting you to normalizeyour blood sugars
for two weeks and then go back to your old diet. My intent is to
demonstrate the value of using insulin, and the value of normalized
blood sugars. We might envision that a mild diabetic still has three
types of beta cells, active, dying, and dead. Myownbeta cells are likely
all of the last variety— dead. I've mentioned it previously, but if I'd
had the kind of treatment upon my diagnosis more than sixty years
ago that I advocate today, I might still have a significant number of
working beta cells. If you have some beta cell function left, you can
probably increase it by normalizing your blood sugars.
If the prospect of injectingyourself horrifies you, don't let it. Many
people assume injections must be painful, but they needn't be. If
you've alreadybeen usinginsulin for years and find the shots painful,
the likelihood is you weretaught to inject improperly.
Insulin: The Basics ofSelf-Injection 251
Ifyouhave type 2diabetes, sooner or lateryou may require insulin in
jections, either temporarily (as during infections) or permanently.
This is nothing to be afraid of, even though many people with long
standing type 2diabetes spend literally years worrying aboutit. I usu
ally teach all my patients how to inject themselves at our first or
second meeting, before there's any urgency. Once theygive themselves
a sample injection of sterile saline (salt water), they find out howeasy
and painless it can be, and they are spared years of anxiety. If you're
anxious about injections, after you read this section please askyour
physician or diabetes educatorto allow youto try a self-administered
injection (without the insulin).
Insulin is usually injected subcutaneously. This means into a layer
of fat under theskin. The regions ofthe body thatare likely to contain
appropriate deposits of fat are illustrated in Figure 16-1. Examine
your body to seeif you have enough fat at the illustratedsitesto com
fortably grab a big hunkbetween your thumband first finger.
Most diabetics are erroneously taught to inject into their thighs in
spite of the obvious: most thighs have inadequate fat for satisfactory
injections. The net result isthat theinjection ends up going intomus
cle instead of fat andthe timing of the insulin issped up inappropri
Fig. 16-1. Potential sitesfor subcuta
neous injections.
To showyou howpainless a shot can be, your teacher should self-
administer a shot to illustrate that no pain is felt. Your teacher should
next give you a shot of saline or "throw" the needle into your skin to
prove thepoint.Now it's time foryou to give yourself aninjection, us
ingasyringe that's already empty or has been partly fiUed for youwith
about 5 "units" of saline.
1. First,with your "nonshooting"hand, grab as big a chunk of skin
plus underlying fat as you can hold comfortably. If you have a
nice roll of fat around your waist, use this site. If not, select an
other site from those illustrated in Figure 16-1. Nearly everyone
has enough subcutaneous buttocks fat to inject there without
grabbing any flesh. Just locate a fatty site by feel. To inject into
your arm, use the topof a chair, theoutside corner of two walls,
or the edge of a doorway to pushthe loose flesh fromthe backof
your arm to a forward position that you can easily seeand reach
with the needle.
2. Hold the syringe like a dart, withthe thumb and first twoor three
fingers of either hand.
3. Nowcomes the most important part. Penetration must be rapid.
Never put the needle against the skin and push. That's the
method still taught in manyhospitals, and it's often painful. If
you can find onlya small amount of flesh to hold, the needle
should piercethe skin at a 45-degree angle, as in Figure 16-2,or
even better, use one of the newinsulin syringeswith a short nee
dle (5/i6 inch). If youcan graba heftyhandful, you should plunge
the needle straight in, perpendicular to the skin surface, or at
Torry Eppridge
Fig. 16-2. Ifyouare skinny, pierce the
skin at a 45-degree angle, orusea
short (Vi6-inch) needle.
Fig. 16-3.Ifyou're chunky, pierce the
skin at anyanglebetween 45 degrees
and 90 degrees.
Insulin: TheBasics of Self-Injection 253
anyangle between 45degrees and 90degrees, as shown in Figure
4. The stroke shouldbegin about 4 inches fromyour target to give
the moving needle a chance to pick up speed. Pretend you're
throwing a dart — but don't let goof the syringe. Move your en
tire forearm and give the wrist a flick at the end of the motion.
You shouldn't get hurt. The needle should penetrate the skin for
its entire length.
5. As soon as it's in, push the plunger all the way down to inject
the fluid. If the demonstration syringe is empty, then don't
bother to push the plunger. Now promptly remove the needle
from the skin.
There's no need to practiceinjecting oranges, as has been taught in
the past. If you're going to practiceanything, you might first practice
"throwing" a syringe, withthe needle cover on, at your skin.
All you need do is experience one rapid stick to realize that speed
makes it painless. Never has it taken more than a moment for me to
get a patient to self-inject. I'vehad grownmen in tears at the prospect
of injecting insulinwhosoondiscover that it'seasy and painless and of
considerable value in treatment. It doesn't demand much skill, and
certainly doesn't require bravery.
In recent years, a number of new insulin syringes have appeared on
the market in the United States. Although they are all sterile, plastic,
and disposable, some are better than others.The important features to
consider are describedbelow. Refer to Figure 16-4, which identifies the
parts of a typical insulin syringe that you might find at your local
The Scale
When selecting a syringe, the printed scale is the most important
feature, because the spacing of the markings determines how accu
rately you can measure a dose. Think Laws of Small Numbers: accu
racyand consistencyof dose are both highlyimportant.
Insulindoses are measured in "units." One unit of our most-rapid-
acting insulin will lower my blood sugar by 60 mg/dl. One unit will
lower the bloodsugar of a 45-pound chUd byabout 160 mg/dl. Some
of myslimadultpatients with mUd type 2diabetes find that 1unit wUl
drop themby80mg/dl. Clearly, an error of onlyV4 unit canmake the
difference betweena normal blood sugar and hypoglycemia for many
of us. MyinsuUn-using patients never inject as much as 8 units in a
single dose. It would therefore beideal to have a long, slender syringe
witha total capacityof 10units andmarkings for every lA unit spaced
farenough apartthat V& unit can beaccurately estimated visuaUy. The
numbers on the scale shouldbeeasy to read. Thelinesshouldbe dark,
but no thicker than Vn unit. Sucha syringe, unfortunately, does not
exist quite yet.
A currently avaUable preferred syringe is Ulustrated below. Note
that the scaleline nearest to the needle is longer than the other lines.
This is the zero line. It overUes the end of the gasketwhen the plunger
ispushedinfuUy. It isnot the 1-unit line. Theupperscale in Figure 16-4
displays whole units; thelower scale shows halfunits.
Gasket Barrel
I 1 _
iTiffWrnjirrn 1111ilrni ih 111il~t°-
Zero Line Scale
Fig. 16-4. Apreferred insulin syringe, calibrated in half-unit increments (enlarged
The Rubber Gasket
This is the dark-colored piece of synthetic rubber at the end of the
plunger nearest the needle. It indicates a given dose by its position
along thescale. Thebestgasket has asurface that's flat andnot conical,
as some are, so that doses can be read without confusion. Note: the
endof thegasket that isnearest theneedle istheendthat shouldbeset
at the dose.
The Needle
The needle should be %-%6 inch long. Longer needles may go too
deeply into thin people. UntU 1996 aU disposable insulin syringes
sold in the United States had V^-inch needles. Syringes with shorter
Insulin: The Basics ofSelf-Injection 255
(5/i6-inch) needles are nowavaUable. With these syringes you usuaUy
neednot "grab a hunk of flesh" or inject at a 45-degree angle unless,
likeme, youhaveveryUttle fat at the injection site. Justthrowit in. Do
not, however, use short needles for intramuscular injection, as de
scribedon page 309.
Needle thickness is specified bygauge number, just as for naUs and
wire. The higher the gauge number, the thinner the needle. With a
verythin gauge, evenpenetrating the skin too slowly may not hurt.
Withtoothin a gauge, theneedle mightbendor breakwhen punctur
ing tough skin. The ideal compromise between thinness and strength
is probably31 gauge, whichis nowwidely avaUable.
The Point
Theneedle pointsof disposable insulin syringes currently soldin the
United States arequitesharp. Advertising that claims special sharpness
for a particularbrand is usuaUy exaggerated.
My technique for filling a syringe with insulin differs from what is
usuaUytaught,but it hastheadvantage ofpreventing thedevelopment
of air bubbles in the syringe. Although it is not harmful to inject air
bubbles belowyour skin, their presence in the syringe interferes with
accurate measurement of smaU doses.
General Technique
This step-by-step approach maybe foUowed foraU clear insuhns. Only
oneinsulinnowon themarketiscloudy. It iscaUed NPHin the United
States and isophane overseas. If youusecloudy insulin, besureto read
thesection that foUows thisonebefore proceeding.
1. Take the cap(s) off your syringe.
2. Drawroomair into the syringe bypidling the plungerbackuntU
the end of the rubber gasket nearest the needleis set at the dose
you intend to inject. If the gasket has a dome or conical shape,
thedose shouldbesetat thewidest part ofthegasket, not at itstip.
3. Puncture the insulin vial with the needleand inject the air into
the vial. This seemingly useless step has a purpose. If youwere
256 Treatment
not to inject air to replace the insulin you withdrew, after many
fiUings a vacuum would eventuaUy develop in the vial, which
would makesubsequent fiUings difficult.
4. Invert the syringe and vialand hold themverticaUy, as shownin
Figure 16-5, then rapidly puU back on the plunger until the bar
rel is filled with insulin weU beyond your dose (e.g., to about 15
units if your dose is to be 5 units).
5. Slowly push the plunger in, stiU holdingverticaUy, until the ap
propriate part of the rubber gasket reaches the desired dose.
6. Continue to hold syringe and vial vertically as you remove the
filled syringe and needle fromthe vial.
Filling a Syringe with Cloudy Insulin
Oneintermediate-acting insuUn (NPH) issoldtodayinvials that contain
a clear Uquid and a gray precipitate. The grayparticles tend to setde
rapidlyfromtheUquid when thevial isleft undisturbed. Theymustbere-
suspendeduniformlyintheUquid immediatelypriortoeveryuse. FaUure
to do this wiU result in inconsistent effects upon blood sugarsfrom one
shot to another. The way to secure a uniform suspension is to shake the
vial. Many years ago, egg white-based vaccines were of a syrupy consis
tencyandtended toform apermanent foam when shaken. This isnot the
case with today's water-based insulins. Yet most textbooks —and even
the American Diabetes Association — still teUnurses and doctors to roU
the vial between the hands and not to shake it. This misinformation is un
fortunate, because wedon't getconsistent results when vials areroUed.
WhenfiUing a syringe witha cloudy insulin, observe the following
procedure to ensurean even suspension.
1-3. Remove cap(s) fromsyringe, drawair into it, and inject the air
into the vial as described in steps 1-3 on pages 255-256.
4. Before drawingout anyinsulin,whUe stiUholding the vial and
syringe in one hand, vigorously shake them back and forth
6-10 times as shown in Figure 16-6. Holding the upward-
pointing syringe and vial vertically, rapidly draw back the
plunger immediately after shaking to fill the syringe with in
sulinweUbeyond your dose. Donot delay, asthe grayparticles
wUl settle very rapidly.
5-6. StiU holding verticaUy, slowly push plunger in until desired
dose is reached, then remove needle and filled syringe from
vial (seesteps 5-6, above).
Insulin: The Basics ofSelf-Injection
Fig. 16-5. Filling the syringe, holding
vial and syringe vertically.
Fig. 16-6. Shaking a vialofcloudy
insulin before drawing outthe dose.
The annual cost of sterUe disposable insulin syringes can beconsider
able, especiaUy if you take multiple daUy injections. You may become
tempted to reuse your syringes, especiaUy if your medical insurance
doesn't reimburse you for the cost. (Many medical insurance poUcies
inthe United States do at least partiaUy cover this expense.) Although
Ihaven't encountered any infection caused by asingle person reusing
his own syringes, I have encountered the problem of polymerization
of insulin.*
Many of mypatients pass through astage when they routinely reuse
their syringes several times, tosave money ortoenable themtotravel
with onlyasmaUsupply. These patients never use the same syringe for
two different types of insulin, so we can't say that one insulin iscon
taminating another. Inevitably, Iget atelephone caU with the message,
*Apolymer isa large molecule made up of identical smaUer molecules bound
258 Treatment
"Myblood sugars are high and I can't get them down." I ask, "Bring
your clear insulin to the phone. Is it crystal clear, like water?" In
evitably the reply is, "No, it's slightly hazy." Insulin that becomes hazy
has been partially deactivated by polymerization and will not ade
quately control blood sugars. This is not found bypeople whodo not
reuse their syringes.* Of course, I advise such patients to immediately
replace all insulin vials, whether long- or short-acting, that have been
used to fill reused syringes. Replacement of the vials always cures the
problem. Naturally, syringes should not subsequently be reused.
What if you encounter a situation whereyou onlyhaveone syringeto
last for a weekand haveno wayof getting newones? Flush the syringe
with air several times after each use to clear out any remaining insulin.
Whenfilling the syringe, do not inject air into the insulin vial (step 3),
and don't inject the excess insulinbackinto the vial (step 5). Just draw
the needle from the vial and squirt the excess into the air.This way, you
won'tcontaminate yourvial with theminute amount of oldinsulin that
may remain in the needle or syringe. If you have a second unused sy
ringe, youcanuseit just to inject air intoyour vials, making certain the
needle does not come into contact with the insulin in the vials.
If for financial reasons youreally must reuse your syringes, the fol
lowing procedure should help minimize contamination with poly
merized insulin. You will, at the minimum, need three syringes, but
four would be better.
• Set aside one syringefor each of the different insulins you may be
using. Put a small piece of adhesive tapeon each syringe, marked
with the abbreviated name of the insulin (see page 269).
• Use your insulin vials for a week without injecting air into them.
Squirt any excess insulin from each filling into a sink or waste-
basket, not back into the vial.
• At the end of a week, remove the plunger from an unused sy
ringe. Stand the vial stopper up on a flat surface and push the
needle of the unused syringe into the stopper of the vial. Within
seconds, the vacuum in the vial will suck in enough air through
the needle to replace the vacuum.
* The reason for this is that the minute amount of insulin remaining in a used
needlewill becomepolymerized (inactivated) within a few hours. If it is injected
back into the vial, it will eventuallyact as a seed for the polymerization of much
of the insulin in the vial.
Insulin: TheBasics of Self-Injection 259
• PuUtheneedle out ofthestopper, reinsert theplunger, andrecap
the "air" syringefor use the next week.
As discussed in Chapter 19, "Intensive Insulin Regimens," you might
have toinject several different insulins at thesame time. For example,
when you arise in the morning, you might inject an ultrarapid in
sulin (e.g., Uspro) to bring down a sUghtiy elevated blood sugar,
then a rapid-acting insulin (regular) to cover your breakfast, then a
long-acting (basal) insulin (i.e., detemir or glargine). Take the most-
rapid-acting (lispro), then the rapid-acting (regular), and last the
long-acting, one injection after another, aU using the same syringe.
You cansafely dothisbecause theinsulin has not hadenough time to
polymerize intheneedle (this takes several hours). Ifyour long-acting
insulin is glargine, however, don't use the same syringe. Just a smaU
amount of Uspro or regular in theneedle mayeventuaUy cause the in
sulin inthevial ofglargine toturncloudy andlose some ofitsactivity.
This is very important.
Don't mix different insulins together, either in thesame syringe or
inthesame vial, as this wiU result inanew insuUn with new, inappro
priate timing. The onlyexception wouldbe to slowdown the action of
regular insulin when dealing with gastroparesis.
Most textbooks and instruction sheets that teach insulin injection or
finger stickingadvisethat the skinshould be "sterilized" with alcohol
before puncturing with a needle. Alcohol wUl not sterilize your skin.
At best it wiU clean off dirt. My patients andI have given miUions of
injections and finger sticks without using alcohol. Noneof us has be
comeinfected as a result. Certainly it'sa sensible ideato cleanoff visi
bledirt first, but youcando this with simple soap and water on the
rare occasions that it may be necessary. I often inject myself right
through myshirtor trousers (but not through thetrouser pockets).
260 Treatment
A Cost-Free Method
I recommend the foUowing cost-free, safe method for disposing of
used syringes. Again, this is contrary to the recommendations of
the ADA.
1. Recap the needle.
2. Putthecapped syringe intoalarge plastic bottle such asthat used
for bleach, bottled water, seltzer, or soft drink. Alternativelyuse a
largecoffee can.
3. When the container is fuU, replace the cover or cap and prevent
its removal by applying duct tape.
4. Put the container into your trash* or take it to your physician,
hospital, or pharmacy for pickup bythe special disposal service
that theymight use.
5. Ifyouareinahotel or restaurant, don'tput used syringes intheir
trash containers unless youfirst put themwith needle recapped
in an opaque plastic bag, sealed with tape or knotted closed.*
Their cleaning people wiU not appreciate seeing loosesyringes.
In an airplane, recap the needle and put the syringe in the trash
bin in the lavatory.
A High-Tech Method
It's nowpossible to melt needles offinsulin syringes. If youare espe
ciaUy conscientious about our environment or would enjoy using a
briUiant technological device, you might try the Disintegrator Plus.
This is a smaU device in a plastic box measuring about 6 x 3Vi x 2xh
inches (15.24 x 8.9x 6.35 cm) and powered bya buUt-in rechargeable
battery. It seUs for$49 at CVS (cvs.com) andfor$79 ifpurchased from
the manufacturer.
To operate thedevice you merely insert theneedle intoa hole and
press ontheactivatingbutton for 3seconds. The needle wiUdisappear,
*Somecommunities forbidthe disposal of suchcontainers in local garbage. It is
wise to contact your garbage coUection department foradvice. You canalso visit
www.safeneedledisposal.org to access a national database of local regulations in
the United States.
f I save plastic grocery bags forthis purpose andalways pack a few empty ones
when I travel.
Insulin: The Basics ofSelf-Injection 261
melted into a tiny metal blob that is stored inside the instrument. You
can then drop the plastic syringeinto your ordinary trash container.
The stored metal blobs can be removed after a month or so byun
screwing a small screwin a hatch on the bottom and pouring them
out. This may take 30 seconds.
Disintegrator Plus is distributed by Perfecta Products, Inc., of
North Lima, Ohio, phone (800) 319-2225. It's fun to use and keeps
needles off our beaches.
Nowadays, most of us will inject through thin clothing (shirts, stock
ings, trousers) when it's inconvenient to undress. This can cause a
problem on the rare occasion that the needle encounters a small blood
vessel. Adrop of blood can appear at the puncture site and stain your
clothing. Finger punctures sometimes bleed more freely than youex
pect, so that upon squeezing you mayget a squirt in the eye, or blood
on your tie, if you're not careful.
The answer to bloodstains on clothing is hydrogen peroxide solu
tion. Hydrogen peroxide isvery inexpensive and is soldin allpharma
cies. Purchase several small bottles. Keep a bottle of peroxide handyat
every location where you measure blood sugars. Carry a small bottle
in your luggage when you travel. Once a bottle has been opened, the
solutionremains stablefor perhaps sixmonths, so you might want to
have a backup bottle available.
You canmake bloodstains disappear very simply without bleaching
the dyes in your clothing. It'sbestif youtreat thestain while the blood
is still wet, as dried blood bleaches veryslowly. If you allowthe blood
to dry, it may take 20 minutes of rubbing to get rid of the stain. Pour
some peroxide on a handkerchief and rub it into the stain. Theperox
ide will foam when it contacts blood. Keep applying and rubbing un
til the stain has vanished.
Many devices have been advertised with the claim that they inject in
sulin "without pain." Since most diabetics have not been taught the
262 Treatment
high-speed painless injection technique described in this chapter,
manyof these special, spring-driven devices aresoldeveryyear. Ifyour
injections arealready painless, it makes Uttle sense to usethem.
Other "painless" devices, caUed jet injectors, use veryprecise con
struction to inject a high-pressure jet of insulin, penetrating the skin
without a needle. These injectors do not require a separate syringe
since theymust be loaded directly withinsulin, using special adapters
that plug into the insuUn vial. Although the concept is veryenticing,
sprayinjectors posesomeproblems. First, they're veryexpensive, cost
ingfrom$300 to $600 intheUnited States. Although thisisa substan
tial initial investment, the cost can be recovered over the course of a
year or two if you're giving yourself lots of injections withdisposable
syringes. Your insurance plan maynot payfor a jet injector but wiU
most likelypayfor disposable syringes.
They'renot as convenient asdisposable syringes because theymust
be taken apart and sterilized in boiling, deionized water everyone to
two weeks. Also, the adapters for the insulin vials sometimes leak
when the vialsare carried in a purse or bag.
You wiU require considerable training and experimentation with
pressure settings in orderto give yourself a properjet injection. This
can delaygettingyour blood sugars normalized. You mayalsoexperi
ence sUghtiy more pain than you wouldwith a speedily injectedshot
froma conventional syringe, and there'sa highincidenceof blackand
blue marks on the skin and minor bleeding and even loss of smaU
amounts of insuUn at the puncture sites.
Despite these drawbacks, jet injectors do have two unique advan
tages, aside from reducing the number of syringes you must dispose
of. Firstis that youwUl require about one-thirdless insulin, since the
shots are better absorbed. Second, if you use fast-acting insuUn to
lower elevated blood sugars, it wiU work even faster. But not faster
than an intramuscular injection (page 307). FinaUy, jet injectors
should not be usedfor longer-acting insulins.
Several manufacturers are advertising "insuUn pens." These are sy
ringes into which smaU cartridges of insulin canbe loaded. Theyare
intended to reUeve you of the burden of carrying a vial of insulin if
youhave to injectaway fromhome.None of thosemarketedas of this
Insulin: The Basics ofSelf-Injection 263
writing canbe set at quarter-unit increments, andonly oneof those
sold in the United States can be set at half-unit increments. Most
therefore cannot provide the fine-tuning ofblood sugars that ourreg
imens require. Stay away from themunless youareveryobese and re
quirelarge doses of insulin. Withlarge doses, an error of Vi or Va unit
is insignificant.
Thecartridges used for insulin pens areless thanone-thirdthesize
of standard insuUn vials and are consequently more convenient to
carryinapocket or purse. You canpuncture thecap withtheneedle of
a standard insulin syringe andslowly draw out insulin. Donot inject
air or reinject insulin into these cartridges.
Important Information About
Various Insulins
Ifyou start using insulin, you ought to understand how itseffects
can be controlled. It can do some remarkable things, but it must
be handled with respect and knowledge. Muchof the information
in this chapter is based upon my experience with my own insulin
needs and with those of mypatients. Asin much of this book, you will
likely note that some statements contradict traditional teachings and
manufacturers' literature.
There are a confusing number of brands and types of insulins being
marketed today —andeven more areontheway. Insulins maybecat
egorized by how long they continue to affect blood sugars after injec
tion. There are most-rapid-acting, rapid-acting, intermediate-acting,
and long-acting insulins. Until recently, the rapid-acting insulins ap
peared clear, like water, and the other insulins appeared cloudy. The
cloudiness is caused byan additive that combines with the insulin to
form particles that slowly dissolve under the skin. Theone remaining
intermediate-acting insulin, called NPH, is modified with an animal
protein caUed protamine. Insulins that contain protamine may stimu
late the immune system tomake antibodies to insulin. These antibod
ies can temporarily bind to some of the insulin, renderingit inactive.
Then, unpredictably, they can release the insulin at a time when it's
not necessarily needed. This effect, althoughsmall,impairs the metic
ulous control of blood sugars that we seek. Protamine can present an-
Important InformationAboutVarious Insulins 265
other, more serious problemifyou ever require coronaryangiography
for the study of arteries that feed your heart (acommon procedure
nowadays). Just before such astudy, you would be given aninjection
of the anticoagulant heparin to prevent theformation of bloodclots.
When theprocedure isover, protamine isinjected intoa blood vessel
to"turn off" the heparin. This can cause severe aUergic reactions, even
death, in a smaU percentage of people who have previously been
treated withinsulin containing protamine. Thus, even if an insulin is
marketed as a"human" insulin, its effects upon antibody production
may besignificant if it contains the animal protein protamine.
As you may guess, I strongly oppose the use ofinsulins containing
protamine. In theUnited States, theonly oneiscaUed NPH(elsewhere
it may be caUed isophane insulin). NPH or mixtures of NPH and
other insulins are widely avaUable and should beavoided. People who
require very smaU doses of insuUn, such as chUdren, may be best
treated with dUuted insulin (page 272) for accurate dose measure
ment. Unfortunately, there isnodUuting fluid made for glargine, one
of our two remaining long-acting insulins.* I therefore amnowreluc
tantly obUged to prescribe three daUy doses of dUuted NPH on rare
occasions. More commonly, however, I'UdUute thelong-acting insulin
detemir with saline, as described onpage 272.
AUst of the insuUns that I consider possibly suitable appears on
page 269.
Thebiological activity of insulin is measured in units. At smaU doses,
2 units of insulin should lower blood sugar exactly twice as much as
1 unit An insulin syringe is therefore graduated in units, and the one
shown in Figure 16-4is alsocaUbrated in half-units. The lines are far
enough apart so that even Va unit can be reasonably estimated. The sy
ringe we recommend is designed for a concentration of 100 units per
*My favorite long- and intermediate-acting insulins, ultralente and lente,
were taken offthemarket in2006 because they were less profitable tothemanu
facturers. DUuting fluids were available for these insulins. Although theAmeri
can Diabetes Association made no protest when these discontinuations were
announced, there iscurrendy anuproar over this intrusion uponpatient care in
the U.K.
266 Treatment
cc, andcandispense up to 3/io cc, or 30units. Theinsulin's strength is
designated U-100, meaning"100 units per cc." Inthe United States and
Canada, this is the onlyinsulin concentration sold, so you need not
specify thestrength when you purchase. Other insulin strengths, such
as U-40 and U-80, are sold in other countries, and the scaleson the sy
ringes inthese countries are designed for these other strengths. Aspe
cial strength, U-500, isavaUable toyour physician intheUnited States,
upon request from the manufacturers, for special appUcations. The sy
ringes for U-40 andU-80 strengths are not sold intheUnited States.
If you travel overseas and happen tolose or misplace your insulin,
you may beunable tosecure the U-100 strength locaUy. You can make
the best of this by purchasing U-40 or U-80 insulin, together with
U-40 or U-80 syringes. You should drawyourusual doses in unitsinto
the newsyringes withthe newinsulin.
Insulinisstable until the expiration dateprintedon thelabel, if refrig
erated. Aslight loss of potency may occur if insulin isstored at room
temperature longer than30-60 days. This isespeciaUy trueof glargine
(Lantus) insuUn, which maylose asignificant amount of potencyafter
60days at room temperature. It isbest stored in a refrigerator.*
Insulin can becomepartiaUy deactivated with or without a change
in its appearance, leading to unexpectedly elevated blood sugars.
When I receive a distress caU froma patient who has had higher than
usualbloodsugars for several days, I aska number of questions in or
der to determine the source of the blood sugar elevation. Have there
been dietary indiscretions? Isthere a possible infection? Or mightthe
insulin be somewhat deactivated, perhaps by reuse of syringes (page
257)? Even sUght cloudiness of aclear insulin isa certain sign of deac
tivation. Sois the appearance of visible clumps within, or a gray pre
cipitate on the waU of, a vial of NPH insuUn (normaUy cloudy) that
wiU not disappear whenit'sshaken. Deactivation of insulin, however,
may not bepossible todistinguish simply bylooking. If dietor infec
tion seems unlikely to be the source of the blood sugar elevation, I
*The manufacturer of Lantus isoverly cautious and recommends that it be dis
carded at 30days after initial use —even if refrigerated. Thisis a very profit-
effective directive.
Important Information About Various Insulins 267
therefore advise my patient to discard aU insulin currently in useand
to utilizefresh vials, evenif the insuUn looksokay.
Here are some simplerules for routine care of your insulin:
• Keep unusedinsulin in arefrigerator untU youare ready to useit
for the first time. Vials in current use maybekept at roomtem
perature for convenience, but Lantus (and possibly detemir and
gluUsine) is best stored in the refrigerator.
• Never aUow insulin to freeze. Even after it thaws out, it may no
longer possess its fuU strength. If yoususpect it mayhavefrozen,
discard it.
• If yourhome reaches temperatures above 85°F (29°C), refriger
ateaU your insulinwhen not in use. If your insulinhasbeen ex
posed to temperatures in excess of 99°F (37°C) for more than 1
day, discard it.
• Do not reuse your insulin syringes (page 257).
• Do not put insulin in prolonged sunUght or in closed, unat
tended motor vehicles, glove compartments,or car trunks. These
areas can become overheated on a sunny day, even in winter. If
you inadvertently leave insulin in a hot vehicle, discard it. This
rule also appUes to blood sugar test strips.
• Donot routinelykeep insulin close to yourbody, asinshirtpockets.
• If youkeepyourcurrently usedinsuUn out of the refrigerator, af
ter it warms to room temperature use a felt-tipped marker to
mark the datewhen the vial was first removed fromthe refriger
ator. Coverthe markingwith clear tape to prevent erasure. Dis
cardaU vialsof glargine, gluUsine, and detemir whenever 30-60
days haveelapsed foUowing the marked date.
• When youinvertyourinsulin vial to fill yoursyringe, observe the
level of insuUn. When the level drops belowthe loweredgeof the
label on the inverted vial, discard the vial. This is especiaUy nec
essarywith normaUycloudy insulins (NPH) becausethe concen
tration of active particles may change asyou use it up.
• If you plan to travel to anarea withwarmclimate where youmay
not be able to refrigerate your insulin, consider a product caUed
Frio, mentioned in Chapter 3,"Your DiabeticTool Kit." This is a
smaU fabric waUet with peUets sewninto the lining. It's avaUable
in five sizesin the United Kingdom and two in the United States.
When thewaUet issoaked inwater for 15 minutes, the peUets wiU
formagel. As the water in the gel slowly evaporates through the
268 Treatment
pores of thewallet, it will keep insulin at asafe temperature with
out recharging for at least 48hours at a surroundingair temper
ature of 100°F (38°C).
It's important for youto know whenyour insulinwUl begin to affect
your blood sugar and whenit will finish working. This informationis
printed on the insert in the insulin package. The published informa
tion, however, maybe inaccuratefor patientson our regimen.The rea
son for this is that we use very small doses of insulin, while most
published data are based upon much larger doses. As a rule, larger
nonphysiologic doses tend to start workingsooner and finish working
later than smaller doses. Furthermore, the action time of an insulin
wiU vary somewhat from one person to another and from smaller to
larger doses. Nevertheless, Table 17-1 is a reasonable guide to the ap
proximate startingand finishing times of the insulins werecommend
when used in physiologic (as opposed to the usual industrial) doses.
Your response may not followa typical pattern, but at least this table
can serve as a starting point.
Insulin action wiU be speeded considerably if you exercise the re
gionof your bodyinto which youinjected. As a consequence, it maybe,
for example, unwise to inject long-acting insulin into your arm on a
day that you lift weights or into your abdomen on a day that you do
In a word, Don't.
Twodifferent insulins should never be mixed — with the single
exception of the specific situation discussed on page 378. Other
than that, mixingof insulinshas no usefulpurpose, eventhough it
is advocatedby the ADA and even though you can purchase mix
tures that are marketed by pharmaceutical companies. Mixing a
long-acting insulin with a rapid-acting one results in an insulin
that no longer has either the long- or rapid-acting properties.
Important Information About Various Insulins
TABLE 17-1
Action time after injection*
Generic name
of insulin
U.S. brand
name Designation
Aspart* A Novolog Most-rapid-
20 minutes 6-7 hours
Lispro* H
Humalog (but assume
Glulisine G Apidra
5 hours)
R Humulin R
Novolin R
Rapid-acting 45 minutes 8-10 hours or
more (but
assume 5 hours)
N Humulin N
Novolin N
2-3 hours 12 hours if
injected in the
morning; 8 hours
if injected at
Detemir* D Levemir Long-acting Slowlyover
4 hours
18 hours if
injected in the
morning; 8-9
hours if injected
Glargine LAN Lantus
at bedtime
*Doses exceeding 7 unitswill usually start sooner, last longer, andact less predictably
than smaller doses. See page 304.
f Aspart isnot quite as rapid in itsaction as lispro or glulisine.
*Canbe diluted for usebychildren (see page 272).
** Doses of NPH that exceed 7 units mayhave a peakof actionat about 8 hours after in
When you're filling in the information on your Glucograf data
sheets, it will be more convenient for you and your doctor if you use
the abbreviated designations shown in the "Abbreviation" column of
Table 17-1 — A, H, G, R, N, D, or LAN — instead of the full names.
270 Treatment
Since it's implied, you also needn't write out the word"units" when
noting insulin doses. Seven units of regular insulinwouldabbreviate
as "7 R," and so on. If you forget these abbreviations, they are printed
in the upper right corner of your Glucograf III data sheet.
If we ignore the differences in timing, 1unit of each of these insulins
will havethe same effect upon blood sugar as 1 unit of anyof the oth
ers, with the striking exception of lispro, glulisine, and aspart. These
insulins are about 50 percent more potent than regular, the only re
maining rapid-acting, true human insulin.
Yes and no. Aspart, lispro, glulisine, detemir, and glargine require a
prescription from your doctor in the United States. NPH and regular
may be purchasedwithout a prescriptionin most states. Local regula
tions are subject to change.
Glargine and detemir, the clear longer-acting insulins, serve a purpose
different from that of the rapid-acting insulins. Indeed, for our regi
menstheyhave but oneprincipal task —to keep bloodsugarfromris
ingwhile fasting (see the discussions of gluconeogenesis and the dawn
phenomenon, pages 92-93). They are our basal insulins. They're not
intended to preventthe bloodsugarriseaftereating. Furthermore, they
are not used to lower a blood sugar that is too high — they work too
slowly for this. Asecondary purpose of longer-acting insulins in mild
type 2 diabetes is to help delay or prevent beta cell burnout. As you'll
seelater, we may use a rapid-acting insulin to cover meals, whether or
not the longer-acting insulins are used to cover the fasting state. Which
insulin to use, and when, depends upon blood sugar profiles.
Important InformationAbout Various Insulins 271
Now, why might we use both the intermediate-acting insulin
(NPH) and the long-acting ones (detemir, glargine)? Won't just one
type or the other suffice? Which one touse depends upon blood sugar
profiles. If your bloodsugartypically rises between noon and bedtime
ondays thatyou skip all your meals, you'll need glargine or detemir on
arising in themorning. We use these andnot NPH inthe daytime be
cause they last longer and will usually carryover until a bit past bed
time. On the other hand, we may useone or, in rare cases, two of these
insulins at bedtime, to cover the overnight fasting state. When one is
needed at bedtime, we usually try the longest-acting first. Glargine
and detemir are especially valuable if the dawn phenomenon is pro
longed or ifyou sleep longer than 8hours. Ifour initial dose ofalong-
acting insulin is not adequate, we may increase it. Sooner or later,
however, we may find that late-morning blood sugars are going too
low, due to the higherdose. We might thenswitch over to NPHat bed
time, to concentrate action during the sleep period. One must be care
ful, however, not to give too much NPH at bedtime, because large
doses may cause bloodsugars to drop in the middle of the night. In
practice, I prescribe intermediate-acting insulin for fewer than 1per
cent of my insulin-using patients. There are other rare uses of NPH
discussed elsewhere in this book.
If you're a type 1diabetic —or a type2 diabetic whois following our
diet and using oral medication and still experiencing bloodsugar in
creases after one or more meals — injecting regular (R), lispro (H),
glulisine (G), or aspart (A) insulin prior to these meals is indicated. By
sheer coincidence, the 5-hour (assumed) minimum action time of
regular corresponds approximately to the time most of us require to
digest fully a mixed meal of protein and carbohydrate, and to experi
ence the final effect of the meal upon blood sugars. Regular insulin
should usually be injected 45 minutes before a meal, so that it starts to
work just as we start to eat.
Thebeta cells of some type2s, however, mayenjoy enoughof a rest
from oneor twosmall doses of glargine or detemirthat theycanproduce
sufficient insulin to cover meals. Since everyone is different, your insulin
regimen must be custom-tailored to normalize your personal glucose
272 Treatment
profile. All this takes more effort onthe partofyour physician than just
theprescription ofone or two daily shots ofalong-acting insulin.
Because of their very rapidaction, lispro and glulisine arealso the
insulins that weuseto lower a highbloodsugar. Since elevated blood
sugars are the cause of the long-term complications of diabetes, we
naturallywanttosee themcome down to normal asfast aspossible. In
Chapter 19, we will teach you how to rapidly get highbloodsugars
down to your target, using Gor H insulin. If your doctor finds that
yourblood sugars are rarely elevated or appear to rapidly dropdown
on their own,thenit maynot benecessaryto useadditional insulinfor
this purpose.
Many type 2 diabetics, mild type 1diabetics, andsmall children with
type 1diabetes require suchsmall doses of injected insulinthat dosage
cannot be measured accurately enough with any of the syringes cur
rentlyon the market. Forsuchpeople, 1unit might lower bloodsugar
bymore than 120 mg/dl (versus only10mg/dlfor a veryobese type 1
or a very obese insulin-requiring type 2 adult). A measurement error
of Va unit would therefore be equivalent to more than 30 mg/dl. To
solvethis problem we dilute the insulin. This is very easy. Your physi
cianor pharmacistcansecure, at no charge, emptysterile insulinvials
from the insulin makers. The manufacturers will also provide, at no
cost, the appropriate diluting fluids for some of the insulins you use.
As if this writing, there are no dilutingfluids for aspart, glargine, or
glulisine insulins, so these cannot be diluted for children. I have re
cently discovered, however, that 0.9percent sterile injectable saline so
lution can be used to dilute detemir insulin.* Thus we again have a
long-acting insulin that can be diluted for small people, and are no
longer obliged to rely on 3 shots of NPHinsulin spread over the day
for those who require dilution.
If your pharmacist is unwilling to performthe dilution for you, ei
ther find a pharmacy with a compounding chemist or do it yourself as
*The saline, in 10, 20, and 50 ml vials with rubber stoppers, is available at any
pharmacyupon the prescription of a physician.
Important Information AboutVarious Insulins 273
1. Have clear instructions from yourphysician as to howmuch in
sulin andhowmuch diluting fluid should be put into avial. If
your doctor writes "dilute 2:1" (say "two to one"), this means 2
parts of diluent, or diluting fluid, for every 1of insulin, and so
on. He may want to give you a few sterile 3 cc syringes* for this
purpose. They will contain about ten times asmuch asthe 25- or
30-unit syringe you use for injections. Using the larger syringe
will speedup the preparation of yourvials.
2. Each vial can hold only 10 cc of fluid. You should write down
how many cc's of diluting fluid and insulin you will need, re
membering that the sum of the two cannot exceed 10 cc. Thus,
if yourdoctor tells youto dilute your insulin 3:1, youmight use
6 cc of diluent and 2 cc of insulin.
3. All diluting fluids should be crystal clear, like water. Make sure
thatthelabel of thediluting fluid youare using specifies that it is
for the insulin youwantto dilute. The diluting fluid for lispro is
the sameasthat for NPH. Regular (R) insulin has its own dilut
ing fluid. As of this writing, none has been made available for
glargine (LAN), aspart (A), orglulisine (G) insulins. Thediluting
fluid for detemir (D) insulinis 0.9 percent saline.
4. Pierce the empty vial with the needle of your 3 cc syringe. Draw
out airto the doseof diluent you wish to transfer(1,2, or 3 cc, et
5. Move the needle and syringeto the diluting fluid vial and inject
the air. Invert syringe and vial and hold vertically while you
slowly withdraw the predetermined amount of fluid. Keep the
tip of the needle near the stopper of the vial to avoid drawing in
air. Be sure to expel anybubbles in the syringe.
6. Inject the diluent into the empty vial from which you took the
air, and withdrawmore airif you will be delivering more fluid.
7. Repeatsteps 4,5, and 6 until the amount of diluent that you had
written down is in the originallyempty vial.
8. Drawanother 1,2, or 3 ccof air(dependingupon how much in
sulin you will be transferring) from the vial you've been filling
with diluent, but this time inject the airinto the insulin vial. In
vert syringeand vial and, holdingvertically, drawout the prede
termined amount of insulin. Keep the tip of the needle near the
Theyshouldbesupplied withrelatively wide-bore (21-23gauge) needles.
274 Treatment
stopper of the vial to avoid drawing in air. (If you're working
withNPH [cloudy] insulin, remember to shake the insulin vial
vigorously6-10 times immediatelybefore withdrawingthedose;
seeFigure 16-6, page 257.)
9. Inject theinsulin into the vial towhich diluent hadbeen added.
10. Repeat steps 8 and9 until thedesignated amount of insulin has
been added to the diluent.
11. Using a permanent-ink felt-tip marker, label the newly diluted
insulinvial with the expiration date that appears on the insulin
vial, the typeof insulin(usethe designation ND,HD, DD, or RD
to indicate that the insulin has been diluted), and the ratio of
diluent to insulin used (2:1, 3:2, 4:1, or whatever it happens to
be). Cover your writing with clear tape to prevent it from rub
bing off.
12. Put the vial of diluted insulinin the refrigeratorfor storage until
its first use.
I've seenmanypeople, includingdoctors, nurses, and pharmacists,
become confusedabout howmuch diluted insulin to inject. With that
in mind, we will run through a coupleof examples to showyou how
simply this canbe computed.
Example 1. Your doctor wants you to inject 2V4 units of an insulin
that has been diluted 1:1. For every 2 parts of liquid in the syringe,
only 1part, or half, is insulin. To get 2lA real units of insulin, you will
have to inject twice as many diluted units (2 x 2V4 = 4V£) as they're
measured on the scaleof the syringe— which is easier to estimate, es
pecially withthe newsyringes that arecalibrated every Vi unit.
Example 2. Your doctor wants you to inject 1V4 units of an insulin
that has been diluted 4:1. This time, for every 5 parts of liquid only 1
part is insulin,sowemust multiply real units by5 to set our dose: 5 x
1V4 = 5% = 6V4 units on the syringe.
I don't really expect mypatients to computethe diluted units they
must take. In the caseof the secondexampleabove,I would askyou to
take 6V4 diluted units. If this werelispro insulin, I'd write "6+ HD" on
your datasheets in the usual doses boxat the top of the form.
ImportantInformation About Various Insulins 275
These three insulins were developed bythree different manufacturers
to overcome regular insulin's inability to rapidly cover fast-acting di
etary carbohydrates. They cannot, however, circumvent the Laws of
Small Numbers relating to large amounts of dietary carbohydrate.
Since fast-acting carbohydrate foods (bread, pasta, fruit, and so on)
usually contain large amounts of carbohydrate, the hazards of using
suchfoods and covering themwithlarge amounts of insulin will still
exist. Furthermore, these foods will still raise blood sugar faster than
thenewinsulins can lower it for people with normal digestion.
There are some applications of theseinsulinsthat the manufactur
ers may not have considered. For instance, if it is inconvenient to take
regular insulin 40-45 minutes before a meal, you can take aspart,
glulisine, or lispro 20 minutes before the meal. They should be fast
enough to cover small amounts of slow-acting carbohydrate without
the 40-45-minute delay. This can bevery valuable when youeat out,
as you will learn in Chapter 19. Also, insulin users who previously
used regular insulin to lower an elevated blood sugar will benefit by
using lispro or glulisine. They will get blood sugar down more rapidly.
This, too, will bediscussed inChapter 19. Note that studies show lispro
toactsomewhat more rapidly than aspart.
Although several insulinsthat do not require injection are and willbe
on the market, noneareof use for theprecise control of bloodsugars
that weseek. Some arebriefly discussed at the endof Chapter 19.
Simple Insulin Regimens
This chapter andthe next describe anumberof specific insulin
regimens. As you read, please refer backto Table 17-1 in the
last chapter for descriptions of the various insulins and their
speed of action — for instance, long-acting insulin will be either
glargine or detemir.
The particular regimen that suitsyouwill dependto aconsiderable
degree upon your blood sugar profiles. Your physician must decide
whether you needlong-acting insulinto cover the fasting state, short-
acting insulinto cover meals, orboth. Ineitherevent,he or shewill re
quireblood sugar profiles and related data, covering asmany days as
he/she designates, priorto everyoffice visit or telephone call for fine-
tuning of doses. Remember that "related data" includes the times of
meals,whether you overate or underate, the times of exercise (includ
ing seemingly inconsequential activity suchas shopping), times and
dosesof blood sugar medications, infectionsor illnesses you may have
had, when and howmany glucose tablets were taken to correct a low
bloodsugar — in short, anything that mighthave affected yourblood
sugar. Bedtime blood sugar readings are especially important infor
mation, because an increase or decrease overnight should most
certainly affect the determination of yourbedtime dosage of longer-
acting insulin.
To give you some examples of howwe might use insulin to bring
yourblood sugar levels into target range, let's consider the following
blood sugar profile scenarios.
Simple Insulin Regimens 277
Let's say you're taking the highest useful dosage ofaninsulin-sensitizing
agent (ISA) at bedtime. Your fasting (i.e., before-breakfast, empty-
stomach) bloodsugars arestill consistently higher than your bedtime
blood sugars. Because of this, you probably require long-acting or
intermediate-acting insulin at bedtime. Before we'd startyouoninsulin,
however, we'd examine your data sheet carefully in order to makecer
tain that you finished your last meal of each day at least 5hours prior
to your bedtime blood sugar measurement. No oneshould be given a
long-acting insulin to cover an overnight blood sugar increase caused
by ameal unless delayed stomach-emptying (Chapter 22) is present.
Forpeople who customarily sleep 8hours or longer, we usually start
witha long-acting insulin; wemay(rarely) start withan intermediate-
acting insulin for people who sleep 7 hours or less. If youlike to sleep
more than 8 hours on weekends, it's wise to use a long-acting insulin
rather than an intermediate-acting one every night, instead of trying
to switch between one and the other.
Because of the dawn phenomenon (page 93), a result of rapid re
moval of insulin from the bloodstream by the liver near the time of
arising in the morning, it's wise to take the bedtime dose of long-
acting insulin no more than 9 hours before the morning dose. The
bedtime insulin will usually appear to have lost much of its action
9 hours after the injection but will start working again after about
3 hours — when the dawn phenomenon ceases.
Estimating the Dose
Your physician may want to use this simple method for estimating
your starting bedtime insulin dose. Generally, 1unit of regular, NPH,
or long-acting insulin* lowers blood sugar 40mg/dl for a 140-pound,
nonpregnant adult whose pancreas produces no insulin. Since your
beta cells may still beproducing some insulin, we'd abide bythe Laws
of Small Numbers and cautiously assume initially that 1 unit of any
*Aspart, glulisine, and lispro areabout 50 percent more potent than the other
insulins. For theother insulins I recommend, as noted in the previous chapter,
except forthespeed with which it acts, aunitof oneinsulin isequivalent toa unit
of any of the others.
278 Treatment
insulinwouldlower you80mg/dl, just sowewouldn'tbringyoudan
gerously lowandriskovernight hypoglycemia.
Wewould then proceedas follows:
First, we'd lookat yourblood sugar profiles. Thefirst number we
want is the minimum overnight blood sugar increase over the past
week. We'd subtract your bedtime blood sugars from your fasting
blood sugars and take the difference from the night with the lowest
rise. For this calculation, bedtime must be at least 5 hours after finish
ing supper. For small children, we accomplish this byasking parents to
get apainless "tushystick" using the Vaculance (see page 69) while the
child is sleeping.
The second number we want is the maximum amount that we'd ex
pect 1 unit of long- or intermediate-acting insulin to lower your
overnight bloodsugar. To get this number, we'd take the maximum
anticipated blood sugar drop from 1 unit. Since our initial conser
vative rule of thumb is that 1 unit of glargine, detemir, or NPH will
lower a 140-pound type 2's blood sugar by80mg/dl, we would divide
140 byyour weight inpounds and then multiply the result by 80 mg/
dl. If your weight is 200 pounds, the equation would look like this:
(140 -5- 200) x 80= 56. So yourinitial estimated bloodsugar dropwill
be 56 mg/dl from 1 unit.
Letus assume, for example, that your lowest overnight bloodsugar
rise inthepast weekwas 73 mg/dl. We'd take 73 mg/dl anddivide it by
thenumber you derived fromtheabove equation, or 56. Your trialbed
timedose of long- or intermediate-acting insulin would be 73 -§- 56=
1.3 units. This isyour starting bedtime dose. Rounding offthedose to
thenearest Va unitgives you Wa units, whichyou can abbreviate onyour
data sheetas 1+ (oneplus) LAN (or Dor N), or just over 1unit.
Fine-Tuning the Dose
Thatwas prettyeasy, but it was onlyastarting point. Most probablythis
dose won't beperfect —likelytooloworpossiblyeven alittle toohigh.
To fine-tune thebedtime insulin, youmerely record bedtime andfast
ing blood sugars for the first few days after starting theinsulin. If the
minimum overnight blood sugar rise was less than 10 mg/dl, you've
hit theproper dose onthefirst try. Iftherise was greater, your physician
may want you toincrease thebedtime dose byas little asVa unit every
third night, until theminimum overnight rise isless than 10 mg/dl.
Even oneovernight hypoglycemic episode canbequitefrightening,
especially ifyou live alone. Such anevent can easily turnyou offtoin-
Simple Insulin Regimens 279
sulin therapy, soit's wise totake some simple precautions to ensure it
doesn't happen. On the night that you take yourfirst shot (andon the
first night of any increase in dosage), set your alarm clock to ring 6
hours after your bedtime injection. When the alarm sounds, measure
your bloodsugar, and correct it to your targetvalue if it's too low(see
Chapter 20). Even onelowblood sugar event suggests that thebedtime
dose should be reduced, or that if you're taking intermediate-acting
(NPH) youshouldpossibly beswitched to a longer-acting insulin.
With the possible exception of growing teenagers, people with de
layed stomach-emptying, or the obese, most of us usually require less
than 8units of long- or intermediate-acting insulin at bedtime. As the
dose of NPH is increased above 7 units, its action tends to peak 6-8
hours after the bedtime injection. This may be a great advantage, be
cause it offsets the dawn phenomenon, or it may cause the problem
just mentioned —hypoglycemia several hours before arising.
Detemir andglargine in doses greater than7units, instead of peak
ing, tend to last longer. This maybe responsible for blood sugars that
are too lowin the late morning, or even in the afternoon. There are at
least twoways to prevent this. First, youcan split the insulin into two
or more approximately equal doses. Theseshould be injected at bed
time, but into different sites. If your required dose is 9 units, you
might inject 4 units into your arm and the other 5 into your abdomen.
You may recall that largedoses are not absorbed with consistent tim
ing or total action, so two or more smaller injections have the advan
tage of making the absorption of both doses more predictable. The
same syringe can be used for the second, third, and so on.
If this method doesn't do the trick for you, your physician mayask
you to inject two separate insulins: one intermediate-acting and the
other long-acting. Hewould customize the relative proportions exper
imentally. We never mixthe twoinsulins together in one syringe.
If your blood sugar rises during the day even though you're taking
maximal doses of one or more ISAs before meals, it's time for you and
your physician to perform another experiment.
This time youwant to determine if meals have causedyour increase
280 Treatment
or if blood sugar increased independently. It's very unusual, by the
way, for fasting blood sugars torise during theday ifyou don'trequire
insulinat bedtime, usually to compensate for the dawnphenomenon
(which, aswe've said, isthetendency inmany diabetics forbloodsug
ars togo upovernight, andperhaps for upto3hours after arising). In
order to determine when and how much your blood sugar is rising
during the day:
• Start your daywitha bloodsugarmeasurement.
• If you're taking anISA inthemorning, continue with yourpres
ent dose.
• Checkbloodsugars again 1hour afterarising.
• Do not eat breakfast or lunch, but plan on a late supper — at
least 12 hours after this secondmorning blood sugar measure
• Duringthe day, continue to check blood sugars approximately
every 4 hours, and certainly 12 hours after the second morning
• If, evenwith a maximal doseof your ISA, your blood sugar rises
more than 10mg/dl during the 12-hour period—without any
drops along the way —you probably should be taking a long-
acting (that is, basal) insulin when youarise in themorning.* We
rarelyusean intermediate-acting insulinin the morning, sinceit
probablywon't last until bedtime.
This dose of basal insulin is calculated the same way we calculated
the bedtime dose in the first scenario. Because fasting twice in one
week is unpleasant, we may tryto waitanotherweek before perform
ingthisexperiment again to see if our basal dose isadequate. Further
experiments in subsequent weeks may benecessary forfine-tuning of
the insulin dose.
Once youtake insulin, it is essential that you and your family be fa
miliarwiththe prevention of hypoglycemia (lowbloodsugar).Tothis
end, youand thosewholive or workwithyoushouldreadChapter 20.
*See page 284 for our introduction to the concepts of basal and bolus insulin
Simple Insulin Regimens 281
It should notbe necessary tomeasure blood sugar every day for the
rest ofyour life ifyou are taking onlylonger-acting insulin as described
in this chapter and you are strictly following our dietary guidelines.
Nevertheless, it'swise toassign oneday every week or two for measur
ingbloodsugaron arising, right before and 2 hours after meals, and
at bedtime, just to make sure that your insulin requirements are not
increasing or decreasing. If any of your blood sugars are consistentiy
10 mg/dl above or belowyourtarget, advise yourphysician.*
It's essential that you also measure blood sugar before andafter ex
ercising. If, in your experience, your blood sugar continues to drop
one or more hours after finishing your exercise, blood sugar should
alsobe checked hourly until it levels off.f
As youshall read inChapter 21, it isimportant whenever yousuffer
an infectious illness to secure daily blood sugar profiles and report
them to your physician.
Many patients and physicians routinely increase the basal morning
dose if before-breakfast blood sugars are repeatedly elevated. This is
thewrong dose to change. It's thebedtime dose that controls fasting
blood sugar, and therefore that dose should be adjusted accordingly.
After fine-tuning of bedtime and,if necessary, morning doses oflong-
acting insulin,your pancreatic betacells mayrecover enoughfunction
eventually to prevent a bloodsugar rise after meals. This frequently
turns out to be the case. If, however, you still routinely experience a
blood sugar riseof more than 15mg/dl 1or 2hours after anymeal, or
more than 10mg/dl 5 hours after anymeal, you'll probably require
premeal injections of a rapid-acting insulin, as described in the next
Weather-Related Changes in Insulin Requirements
Some people experience a sudden decline in their insulin require
mentswhen a longperiodof cool weather (e.g., winter) isabruptlyin
terrupted by significantly warmer weather. This phenomenon can be
*See page299if you've forgotten howwearrive at a bloodsugartarget.
*Insulin users mustalways check blood sugar before they drive andhourlywhile
driving. Ditto for operating potentially dangerous machines. Scuba divers
shouldprobably checkbloodsugarsevery 20-30 minutes.
282 Treatment
recognized byblood sugar well below target when the weather sud
denly becomes warmer. Insuch individuals, insulin requirements will
rise aswinter occurs anddropinthe summer.* Thereason for this ef
fect is speculative, but may relate to the increased dilation of periph
eral blood vessels during warm weather and resultant increased
delivery of glucose and insulin to peripheral tissues. Whatever the
cause, keep careful track of your blood sugar whenever the weather
warms suddenly, since potentially severe hypoglycemia can result if
insulindosages are not adjusted.
Air Travel Across Time Zones
Long-distance travel that requires youtoshiftyourclockby2hoursor
less shouldn't have a major effect upon your dosing of ISAs or basal
insulins covering the fasting state. It should certainly have no effect
upon the use of fast-acting insulin or insulin-sensitizing agents in
tended to cover meals. A problem does arise when travel shifts the
time frame by 3 or more hours and you're taking different doses of
long-acting medication in the morningand at bedtime. Thesituation
becomes particularlycomplexif youtravel halfway around the world,
so that dayand night are reversed.
When the time shift amounts to 2 hours or less, you need only take
your morning medication upon arising in the morningand your bed
time medication at bedtime. One solution to handling larger time
shiftsis to effect a gradualtransition, using3-hour intervals over a pe
riod of days. Todo this, you must keeptrack of the time"back home."
If, for example,you'retraveling east,sothat the time backhome is ear
lier,on your first dayaway youwouldtakeboth of your doses3 hours
later on the "back home" clock. On the second day, you would take
them 6 hours later, and so on. Thus, if your newlocation to the east of
home is in a time zone 6 hours later than it was at home, it would take
you 2 days to achieve a full transition.You woulddo just the opposite
whentravelingwest. Thisprocedurecanbeinconvenient becauseit re
quiresthat youset an alarmclock for absurdhours just to takean in
sulin shot or a pill — and then, you hope, goback to sleep.
Several of mypatients routinelysave themselves this kind of annoy
ancewhentheytravel. Attheir destinations, theycontinueto taketheir
*Some diabetics who alsohavethe diseaselupus erythematosus may experience
just the opposite — lower insulin requirementsin cold weather and higher re
quirements in warm weather.
Simple Insulin Regimens 283
morningdose whentheyarise in the morning andtheirbedtimedose
when they go to bed. They check their blood sugars every 2 hours
while awake andlower them, if toohigh, using the methoddescribed
in Chapter 19. If their bloodsugars drop toolow, they raise them us
ing themethod described inChapter 20. Frankly, this istheapproach
I use myself. Neither I nor my patients have gotten into trouble this
way. This carefree approach can cause problems if the bedtime dose is
considerably different from the morning dose. If this is the case, the
gradual transition of 3hours perday is certainly safer.
Splitting Larger Doses of Insulin
My patientsand I haveobserved that aslarger doses of insulin are in
jected, the effects upon blood sugar become less predictable. This is
duein part to day-to-day variations in absorption of large injections.
After some trial and error, I arrived at a cutoff point of 7 units asthe
largest singleinjection I would want an adult to take (smaller for chil
dren). Therefore, if an insulin-resistant patient requires 20 units of
glargine at bedtime, I askhim to take 3 separate injections in 3 sepa
rate sites of 7 units, 7 units, and6 units,all usingthe samesyringe.
Intensive Insulin Regimens
Alltype 1diabetics but themildest shouldbetreated withrapid-
acting insulin before meals as well as long-acting insulin in
the morning and at bedtime to cover the fasting state. This
roughly mimicsthe way that anondiabetic's body releases insulin to
maintainnormalbloodsugars. Generally, the nondiabetic body when
fasting has a constant, relatively lowlevel of insulin in the blood
stream. This is the baseline, or basal, insulin level to prevent gluco-
neogenesis, the conversion of protein stores (muscles, vital organs)
into glucose. Without it, theywould"melt into sugar water," asthe an
cients observed when diabetes was first described in writing.
During the fasting state (sleeping, between meals), the pancreas
storesthe insulin it creates in preparation forthe next time the body is
exposed to food, while maintaining the lowbasal release rate. Upon
eating and for the first 5 or so hours thereafter, the body receives
what's known asa bolus of insulin— a greater rateof release — until
the glucose derived from meals is stored in the tissues (Figure 1-2,
page 45).As you mayrecall from Chapter 6, the body has counterreg-
ulatory hormones that keep blood sugar from dropping too low so
that one doesn't becomehypoglycemic. So for those of us who make
little or no insulin, essentially what we're trying to do with rapid- and
long-acting (and, in some cases, intermediate-acting) insulins is to
create a rough approximation of asteady basal rate and an appropri
ate bolus rate.
If you are a type 2 diabetic and preprandial (before-meal) use
of ISAs does not prevent your blood sugars from routinely increas
ingby morethan 10 mg/dl at anytime priorto the next meal, it'sprob-
Intensive Insulin Regimens 285
ably time for you to use a rapid-acting insulin —lispro (H), aspart
(A), glulisine (G),or regular (R) —before meals.*
Much of this chapter consists of guidelines for computing insulin
timing and doses invarious situations. Theyare essentiallyprettysim
ple calculations, and your physician or health care provider can and
indeed should make themforyou. I have included themherefor sev
eral reasons. First, you should understand the information that goes
into customizing a dose ofinsulin, so that you know there's nomys
teryinvolved. Second, ifyouunderstand howthese calculations work,
you can also more clearly seewhat incorrect insulin doses look like,
and, we hope, avoid them. Finally, despite the dramatic findings ofthe
Diabetes Control andComplication Trial, many physicians andhealth
care professionals arestill underthefalse impression that normalized
blood sugars are dangerous or impractical or impossible. My hope is
that by providing these calculations, I can help you help your health
care provider provide you with better health care.
Ifyou're not the"math type," you can certainlyskip thecalculations,
but do not skip the entire chapter. Herein lies importantinformation
aboutadjusting yourinsulin dosages or timing toaccommodate com
mon variations in your daily routine, suchas eatingout, and howto
adjust your insulin if you skipa meal or have a snack. (Later in this
chapter, youwill learn why I rarely advocate snacking.)
Theuse of rapid-acting insulin priortoevery meal or snack may help
to preserve the function of anybetacells that youmaystillhave. Nev
ertheless, you might not feel terribly enthusiastic about multiple daily
injections. It'spossible, however, thatyoumay only require insulin be
fore some meals and not others. Several of my patients, for example,
maintainnormal bloodsugars byinjecting rapid-acting insulinbefore
*Clinical trials have shown aspart andglulisine to have virtually the same po
tency as lispro but somewhat slower timing of action, and more rapid action
than regular (R) insulin. I have tried all of the new analog insulins and have
elected to useregular and glargine for mypersonal bolusand basal insulins, re
spectively, andlispro forlowering elevated blood sugars.
286 Treatment
breakfast andsupperandtaking anISA2hoursbefore lunch.Onepa
tient injects before breakfast andsupper, andhas no medication be
fore the small lunch she eats prior to her workout at the gym. The
ultimate determinant of when you require preprandial rapid-acting
insulin isyour glucose profile. Ifblood sugar remains constant before
andafter every meal except supper, thenyouneed rapid-acting insulin
You may recall, from our discussion of the dawn phenomenon in
Chapter 6, that bothyour own andinjected insulins appear to beless
effective whenyou wake up in the morning. This is why virtually all
the people I've seen who require anypremeal bolus insulin must at
least have a dose before breakfast.
Please reread the section entitled "Lispro, Aspart, and Glulisine: New
Ultrarapid Insulins," on page275.
Clearly, when compared to regular insulin, lispro has both advan
tages and disadvantages. Figure 19-1 illustrates the reasonfor a minor
dilemma. As you can see, Humalog, or lispro, has a high early peak
level in the blood, and then after 2 hours its level drops below that of
regular. Attempting to match this peak with the action of carbohy
drate upon blood sugar is verydifficult for several reasons. I won't go
into them all, but consider the following:
• The timingand shapeof the peakwill varyfromone injectionto
the next.
• Theywill alsovarywith the sizeof the dose.
• The appearanceof carbohydrate in the blood willvaryover time
and from meal to meal.
• The flatter peak of regular insulin is easierto match with slow-
actingcarbohydrate than isthe sharppeakof H, A,or Gwith ei
ther slow- or fast-acting carbohydrate.
On the other hand, for most of us, regular must be injected about
45 minutes prior to a mealin order to start workingas the meal starts
to raise blood sugar. Lispro will start workingabout 20 minutes after
injection. This short time interval makes for great convenience if you
Intensive InsulinRegimens
•^ 2 5-
y 2.0 q
I 1.5 4
I 1.0 •:
| 0.5 A
w 0.0 J
Humulin R (n=10)
Humalog (n=10)
(Mean Dose 15.4 U)
i •'• 'Ii•••1111•11111•111111111111111111111111111!
0 60 120 180 240 300 360 420 480
Time (minutes)
'Baseline insulin concentration was maintained by infusion of0.2mll/min/kg human insulin.
Fig. 19-1. Serum insulin levels and action times ofrapid-acting insulins: Humulin
R(regular) versus Humalog (lispro). Note that the "industrial-sized" mean dose,
15.4 units, isfarlarger than the physiologic doses recommended inthis text.
don't know precisely when your meal will beserved, as when dining
out (see below). Withthisinmind, I usually recommend that patients
cover meals withregular whentime permits, but takelispro whentime
is tight. I will usually, therefore, refer to regular as the premeal bolus
insulin. This does not rule outthe use of lispro, aspart, orglulisine, or
lispro plus regular, for situations tobediscussed inafew pages.
There are yet additional complexities to using the three analog in
sulins.* First of all, their effect uponblood sugar is not as consistent,
at least for me and my patients, as that of regular. Second, as men
tioned earlier, these are, in our experience, 50 percent more potent
thanregular, so that their doses must be onlytwo-thirds the dose of
regular for thesame net effect upon blood sugar.
From here on, for the sake of brevity, I'll usually refer only tolispro
whendiscussing the most-rapid-acting insulins.
*Onlytwoof the insulins wediscuss (regular and NPH) have the samemolecu
lar structure as human insulin. All the others have slightly different structures
and aretherefore called "analog" insulins, not "human"insulins.
288 Treatment
Our goal is to minimize or totally prevent anybloodsugar increase
during or after meals. To achieve this, you must take your shot far
enoughin advance so that the insulin begins to lower bloodsugar as
yourfood starts to increase blood sugar. Yet you should not take it so
far aheadof the meal that blood sugar drops faster than digestioncan
keep upwithit.The best time toinject regular, formost of us,isabout
45 minutesbefore eating. The most commonexception wouldoccur
if you have gastroparesis, or delayed stomach-emptying. Our ap
proaches to the diagnosis of this condition andto appropriate timing
of preprandial insulin ifyouhave it aredescribed in Chapter 22.
Determining When to Inject
The following experiment should be useful in determining howlong
before a meal youshould inject your regular insulin. This test canbe
conclusive onlyifyourstartingbloodsugar isnearnormal—perhaps
below 140 mg/dl and level for at least the prior 2 hours.
First, inject regular insulin 45minutes before your planned meal
time. Now, measure blood sugars25,30,35,40,45 minutes, and so on
after the shot.
The pointintime when your blood sugar has dropped 5mg/dl de
termineswhenyoushouldstart eating. If this point occursat 25min
utes, don't evenbother to measure further, just start to eat. If no drop
is seen at 45 minutes, then delay the meal and continue checking
blood sugar every 5minutes until you see atleast a5mg/dl drop. Then
begin your meal. It shouldn't be necessary to repeat this experiment,
unless your preprandial dose of regular is changed by50 percent or
more at some future date.
Ifyourstarting blood sugar ishigher than140 mg/dl when youper
form this experiment, the lack of precision in blood sugar measure
ment andinsulin sensitivity maybe greater thanthe5mg/dl dropthat
we're looking for. Just put offtheexperiment untilyourbloodsugaris
nearer to normal. In the meantime, assume the 45-minute guideline.
Is There Room for Error?
Suppose after performing the above experiment you find that your
regular insulin should beinjected 45 minutes before eating —which
Intensive Insulin Regimens 289
is the case for most ofus. Howfar off can you be without getting into
Eating 5minutes earlyorlate makes nosignificant difference. Ifyou
eat 10 minutes toosoon, your blood sugar may rise during the meal,
butit probably will return toits starting point by the time we assume
the regular finishes acting, about 5hours after injecting. This isnotse
rious, especially if it occurs only occasionally. If blood sugars go up
significantlywith every meal over many years, you would probably be
at risk for long-term complications of diabetes. If you eat 15 or 20
minutes too soon, your blood sugar may go so high (say 180 mg/dl)
thatyoubecome slighdyresistant totheinjected insulin. Ifthisoccurs,
your blood sugar will not drop all theway to thepremeal level when
the regular finishes its action. Ifit happens often, your risk for devel
opingthe long-termcomplications of diabetes will increase.
What ifyou delayyour meal by 10 or 15 minutes beyond the proper
time after your shot? Now you're asking for trouble! Regular starts to
work slowly, but its effect on blood sugar accelerates over the first 2
hours or so. Even a delay of 10 minutes can send your blood sugar
dropping more rapidly than a low-carbohydrate meal can raise it.
This, of course, can be hazardous.
Part of the pleasure of eating out is having someone else serve you
something you can't make at home, but the difficulty for the insulin-
taking diabetic isthatyou're served ontheir schedule, notyours. Host
esses, restaurants, and airlines —as well intentioned as they may
be—rarely serve you at the time they promise. For nondiabetics,
waiting maybeannoying. For those ofus who are diabetic, annoyance
is compounded withdanger. When planning your premeal bolusin
sulin shot, you cannot afford to rely on the word of your hostess,
waiter, or airline staff. I've been taking premeal bolus regular insulin
for more than thirty-five years and have been "burned" more times
thanI care tocount. Nowthatwe have lispro, I inject adose when I see
the waiter approaching my table with the first course. If I suspect that
themain course will be delayed, I'll split mydose in halfand take the
second half when the waiter arrives with the main course. You should
290 Treatment
do the same. A transient blood sugar elevation is a small price to pay
for the assurance that youwill not experience severe hypoglycemia be
cause the meal was delayed. If you eat a low-carbohydrate meal very
slowly, even a transient blood sugar increase can beavoided.
As lispro is 50 percent more potent than regular, its dose should
only betwo-thirds thedose of regular for the same meal.
Nowadays, most airlines serve meals only onoverseas flights. Unless
you are traveling first class, or possibly business class, you will proba
bly have no choice as to meal content. It is therefore wise to bring
along yourown food —or at least theprotein portion,such ascanned
fish or meat or even some cheese. Usually you will be served slow-
acting carbohydrate intheform of salad or vegetables. Again, this isa
timefor using a most-rapid-acting insulin such as lispro 0-20minutes
before you begin to eat.
By the way, never order "diabetic" meals when traveling by air. As
of this writing, airlines are still serving as "diabetic" meals a high-
carbohydrate diet loaded with simple sugars. Thesalads inthese meals
may even contain fruit. My trick is to preorder "seafood" or even
kosher meals when I reserve my flight. This ensures that I get reason
able portions of protein. Unfortunately, many airlines do not serve
seafood for breakfast. On airlines that serve nothing but drinks and a
bagof peanuts, youmayactually be betteroff. You can packyour own
brown bag breakfast or lunch, stick to your diet, and time exactly
whenyou're going to take yourshot and eat your food.
One warning: If you have gastroparesis, never inject a most-rapid-
acting insulinfor a meal, as it will work faster than your stomachcan
emptythe food. Use regular insulin.
Must Meals Be Eaten at the Same Time Every Day?
Eversince the introduction of long-acting insulin in the late 1930s,di
abetics have been advised that they must have meals and snacks at the
same times everyday. Thisveryinconvenient rule still appears in cur
rent literaturedescribing thetreatment of diabetes. Prior to our useof
lowdoses of long- or intermediate-acting insulins to cover the fasting
state, most physicians prescribed 1or 2large daily doses of long-acting
insulin to cover both the fasting state and meals. (Most still do.) Such
regimens never succeed in controlling bloodsugars, and hypoglycemia
Intensive Insulin Regimens 291
is an ever-present threat. Patients are told to eat meals and several
snacks at exactly the right times, to offset the continuous blood sugar
dropcaused bythe long-acting insulin.
But if, as outlined in this chapter, we now cover our meals with
rapid-acting insulin, we're free toeat whenever we want, provided we
take ourshot beforehand. We can also skip a meal if we skip theshot.
When I was in medical training and worked 36-hour shifts, I some
times skipped breakfast andate lunch at 3a.m. Onsome days I didnot
eat at all. This worked out fine because I followed the flexible insulin
regimen described here.
What If You Forget to Take
Your Regular 45 Minutes Before Eating?
Ifit's nowless than 15 minutes before apredetermined mealtime (e.g.,
your lunch break at work), take lispro instead of regular.
Ifyou ate your meal after forgetting your regular insulin, take lispro
instead —immediately—but don't forget that theamount of lispro
should beonly two-thirds your usual dose of regular.
We know that for type 1 diabetics who make no insulin at all, 1 unit of
regular insulin usuallylowers bloodsugar 40 mg/dl ina140-poundadult.
We also know that 1gram of carbohydrate raises blood sugar 5 mg/dl.
Thus 1unitregular usuallycovers 8grams ofcarbohydrate. We also know
that 1unitofregular insulin covers approximately Wi ounces ofprotein.
There arevariables, however. These figures applyonlytopeople who
produce none of their own insulin and who are not insulin-resistant.
Doses must betailored totheindividual, soif you're obese, pregnant,
or a growing child, you may require more insulin than these guide
lines suggest. Ontheother hand, if your beta cells are still producing
some insulin, you may need considerably less insulinthan indicated
here. I have patients who require onlyone-quarter ofthese amounts of
Another variable in figuring a proper dose of regular is our old
friend the dawn phenomenon. The regular insulin you inject before
eating will be perhaps 20 percent less effective at breakfast than at
other meals, eventhough it comesfromthe samevial.
292 Treatment
The biggest factor is, ofcourse, what you eat. Since we cannot know
exactly how regular insulin will affect you until you begin to use it,
your initial trial doses before meals must be based upon your precisely
formulated meal plan. With that, we can make areasonably safe initial
estimateof howmuch insulinyou're likely to need.
It's noteasyfor your physician tobalance outall these variables and
come up with just the right doses of regular insulin on the first at
tempt. Because of this, we try, for safety's sake, tounderestimate your
insulin needs initially, andthen graduallytoincrease your preprandial
doses after checking subsequent blood sugar profiles. This is yet an
other example oftheLaws ofSmall Numbers inaction. Because ofthe
complexity ofthis task, let us examine howyour physician might pro
ceed with two very different scenarios.
You're a type 1diabetic and are switching to our regimen from an out
dated regimen ofl or2 large daily doses ofintermediate- orlong-acting
insulin. Remember that many type 2 diabetics eventually lose nearly all
beta cell function andthen, in effect, have type 1diabetes. So this scenario
would applyto these people too.
Assume that the meal plan you negotiated with your physician is
the following:
Breakfast: 6 grams carbohydrate, 3 ounces protein
Lunch: 12gramscarbohydrate, 4V£ ounces protein
Supper: 12grams carbohydrate, 6ounces protein
Because wewant to play it safe andstaywiththe lowest possible in
sulin doses, wewillfor the moment ignoreanyeffect of the dawnphe
nomenon upon your breakfast dose, as well asthepossibilityofinsulin
resistance dueto obesity. Our approximate calculations, based on the
numbers mentioned above for a 140-poundadult, are as follows:
To cover carbohydrate: number of grams •*- 8 = units of regular
To cover protein: number of ounces -s- 1.5 = units of regular
Intensive Insulin Regimens 293
• 6 grams of carbohydrate * 8 = 3A unit of regular insulin (which
you'd note on your data sheet as 1~ [1 minus] R)*
• 3ounces of protein•*- 1%= 2 units of regular (2R)
• Total trial dose for breakfast will be2% units of regular (3~ R)
• 12 grams of carbohydrate v8=l'/2 units of regular (1 Vz R)
• 4Vi ounces of protein -s- Wi = 3 units of regular (3 R)
• Total trial dose = 4Vi units of regular {iVi R)
• 12 gramsof carbohydrate -4-8=1% units of regular (1% R)
• 6 ounces protein + Vh= 4 units of regular (4 R)
• Total trial dose = 5% units of regular (5!/2 R)
Your physician will probably want to lower these doses if yourpan
creas is making any insulin (as shown either by his or her educated
guess or bythe C-peptidetest, page 54).
It's virtually certain thatyour trial doses will beabit toohigh or too
low. In other words, your blood sugars may either rise or drop after
some or all of these meals. It is most likely, however, that your post
prandial blood sugars will not be dangerously low, unless you have
gastroparesis. If you're insulin-resistant, you will likely need more in
sulin on the second try.
Both you andyour physician will want toget your blood sugars into
line asrapidly aspossible. So you'll probably beasked tofax, phone, or
bring inyour blood sugar profiles during the second day (and perhaps
subsequent days) of this intensive insulin regimen for fine-tuning of
doses. Remember that the important blood sugar measurements for
fine-tuning your doses of premeal insulin are 5 hours after each dose
of regular, aspart, glulisine, or lispro, as we assume this is the time it
takes for the insulin to finish working. Let's assume that on the first
dayyour blood sugar profilelookedlike this:
*Note thatwe use thesymbols +and" toindicate thata dose isjustabove or just
belowthe nearest whole unit on asyringe scale. So1" means 3A unit and 3+ means
3V4 units.
294 Treatment
5 hours after breakfast:increased70 mg/dl
5 hours afterlunch: decreased 20mg/dl
5hours after supper: increased 25mg/dl
Clearly, our initial insulin doses were abit off and require adjust
mentto prevent further increases ordecreases of more than 10 mg/dl.
These changes are easy, if you remember that for most 140-pound
adults who makeno insulin (type 1diabetics), 1unit of regular lowers
blood sugar by 40 mg/dl. If you weigh 100 pounds, 1unit of regular
will lower you about 56 mg/dl, or(140 +100) x40 mg/dl. If you weigh
180 pounds, 1unitof regular will lower you about 30 mg/dl, or(140 +
180) x 40 mg/dl. We will assume, for this exercise, that your weight is
close enough to 140 pounds to use the 40 mg/dl drop from 1 unit
of regular. Type 2 diabetics might dobetter by using Table 19-1, on
page 303.
Now let's look again at the hypothetical blood sugar profiles and
workout thechanges in preprandial regular thatwill benecessary:
Change in dose
Blood sugar Change * rounded off to
Meal change 40mg/dl nearest V* unit
Breakfast +70 mg/dl +1.75 +13/4R
Lunch -20 mg/dl -0.5 -V*R
Supper +25 mg/dl +0.625 +V6R
We now fine-tune our premeal bolus of regular insulin by making
the above changes to the original trial doses.
Meal Trial dose Change New dose
Breakfast 23/4R +1% 4V*R
Lunch 4V£R -Vi 4R
Supper 5V4R +V4 6R
That was pretty easy. Remember, however, that thecontent of your
meals, in terms of grams of carbohydrate andounces of protein, must
bekept constant from oneday to thenext, because your insulin doses
will not be changing every day. If you're consistently hungry after a
particular meal, you can increase theamount of protein at that meal,
but youmust thenhave the extra protein every day. When you raise
IntensiveInsulin Regimens 295
the protein portion of your meal, you look at your blood sugar profiles
(or your physician does) to see how much your blood sugar goes up,
and increase your dose of regular insulin for that meal accordingly. Do
not increase your carbohydrates beyond 6 grams for breakfast, 12
grams for lunch,and 12 grams for supper — the Laws of Small Num
bers dictate that the resultant rise andrequirement for excess insulin
will cause real problems with your blood sugar normalization at
Scenario Two
You have type 2 diabetes and are following our diet. You've been taking
an ISA in the morning and/or at bedtime. Your blood sugars are fine
when you skip meals, butthey go up after meals, even with the maximal
doses ofyourISA.
Since you're not atype 1diabetic and are making some insulin of
your own, we cannot use the simple rules that apply to those who
make essentially noinsulin. Wehave toassume that your beta cells still
make a portion of theinsulin needed to cover your meals, yet wedo
not knowthe magnitudeof that portion. Furthermore, wedon't know
howmuchyour insulin resistance will affect your injected insulin re
quirements. Sowe see howmuchameal will raise yourbloodsugars
without premeal bolus regular insulin. We then use this blood sugar
increase as a guide for the doses you will be needing. We do not use
this method withtype Isbecause their blood sugars mightgo sohigh
without insulin as to cause the dangerous condition known as keto
Further fine-tuning of preprandial regular insulin might be per
formed by reviewing your blood sugar profiles over aweek. If you've
been taking apremeal ISA, as assumed inthis scenario, you probably
have already collected blood sugar profiles that show howmuchyour
blood sugar increases after each meal. Ifthese profiles cover only 1day,
okay. If theycover aweek, better. Wewant to start youwiththelowest
reasonable insulin doses, sowepickthesmallest bloodsugar increases
thatwe can find for each meal, and then adjust your preprandial in
sulin accordingly. To find theincrease before you begin taking regular,
subtract the preprandial blood sugar from the 3-hour postprandial
blood sugar measurement (we wait 3 hours to allowthe effect of the
meal to be nearits highest level).
On pages 277-278, weshowed youhowto compute astarting dose
of long- orintermediate-acting insulin tocover overnight blood sugar
296 Treatment
rises. We can use the samesimple formula to calculate initial doses of
regular insulin to cover meals. But for safety's sake, and to obeythe
Laws of Small Numbers, we're deliberately going to keep the trial
doses on the low side.We'll use as a guide the blood sugar data you
collected while you were taking your ISA, even though we may dis
continue the ISAsometime after you startusingpremeal bolus regular
To finish this example, let us assume that your3-hourpostprandial
increases in bloodsugar over the past weekcanbe summarized as fol
Smallest increase afterbreakfast: 105 mg/dl
Smallest increase afterlunch: 17mg/dl
Smallest increase aftersupper: 85mg/dl
Nowwe must estimate the premeal bolus doses of regular insulin
that would approximately offset these increases. You may remember
that our preliminary formula in estimating trial doses of glargine in
sulin is that 1unit of insulin will lower a 140-pound, insulin-requiring
type2diabetic's blood sugar by80 mg/dl. Your physician maywantto
be even more conservative and assume that 1 unit will lower your
bloodsugar by 90mg/dl. Wenowonlyneed to divide the above post
prandial blood sugar increases by 90 to get the trial doses of premeal
bolus regular insulin, as in the following table:
Rounded off to
Blood sugar Increase * nearest V* unit for
Meal increase 90mg/dl trial dose of R
Breakfast 105 mg/dl 1.17 MR
Lunch 17mg/dl 0.18 V*R
Supper 85 mg/dl 0.94 1R
As in the previous scenario, youwill need to take periodic blood
sugar measurements to monitor the effect of the insulin. If after one
day on thetrial doses of premeal bolus regular insulin, yourpostpran
dial blood sugars still goup by more than 10 mg/dl at 5 hours, your
physician mayask you to increase the appropriate preprandial doses
by V4 unit. (Note that we now look at 5-hour blood sugars instead of
3-hour values, becausewe assume injected regular insulin requires 5
hours to finish working.) If your postprandial blood sugar elevations
IntensiveInsulin Regimens 297
hardly respond to the Vfc-unit increase, your physician maychoose 1-
unit increases. We rarely increase an initial preprandial dose in steps
greater than 1 unit because of the danger of hypoglycemia.
The above trial-and-error procedure should be repeated untilyour
5-hourpostprandial blood sugars donot consistently change from the
preprandial values by more than 10 mg/dl up or down. This all as
sumes that the carbohydrate and protein contents of your meals re
main constant.
If you've ever been on oneof theconventional regimens that utilizes 1
or 2large daily doses of longer-acting insulin, you're probably familiar
with mandatorysnacks. These are required, usually midway between
meals and at bedtime, in the hopesof offsetting the continuous blood
sugar-lowering effect of large amounts of insulin, hopefully prevent
ing dailyepisodesof hypoglycemia.
Our regimen, as you know, uses suchlowdoses of glargine or de-
temir insulins that blood sugars tend to remain level during the fast
ingstate. With our regimen, there is no need for mandatory snacks!
This does not mean that youmust wait until the next meal before eat
ing if you're hungry. Theoretically, you can eat a snack almost any
time, provided that you cover it with regularinsulin,just asyou would
a meal. There are, however, some guidelines to remember.
Snacking Guidelines
Tryto avoid snacks during the initial fine-tuning stage of your insulin
doses. This is especiallytrue of bedtime snacks. Snacks and their doses
of regular insulin canconfuse theissue ofwhat caused what change in
blood sugar. If, for example, you wake up with a high or low fasting
blood sugar, did the problemoriginate in your bedtime dose of inter
mediate- or long-acting insulin, or inthe dose of regular that youtook
for the snack?
Anytime yousnack, try to wait until your prior meal has been fully
digested, and the dose of regular insulin for that meal has run its
course, about 5 hours after the preprandial regular. Suppose youwere
to eat a snack 2 hours after a meal and then were to check your blood
sugar 5hoursafter the regular you took to cover the snack; youwould
have no wayof tellingwhether it wasthe meal or the snack, and the re-
298 Treatment
spective doses of regular insulin, thatwere responsible for anyincrease
or decrease in your blood sugar.
If yousnack, don't eat "snack food." Try to snack on a food such as
asingle serving of sugar-free Jell-0 gelatin (without maltodextrin) or
three small sheetsof toastednori, that is, something that will not sig
nificantly affect your blood sugar and will not have tobecovered with
insulin. Most snacksother than these muddy the waterswhen you're
trying to analyze data. If you really make no insulin you shouldn't
snack, because yourbloodsugar depends almost entirely on what you
eat andinject. Snacking interferes with meticulous bloodsugar con
trol. Full type Is who do snack will have to refrain from correcting
high blood sugars until 5hours after thebolus injection of regular or
lispro for the snack. If you make some insulin and your routine in
jected doses have been fine-tuned, blood sugar corrections after a
snack maynot beneeded, as youmaybeable to make enough insulin
to prevent slighdy elevated blood sugars (or "turn off" insulin pro
ductionif bloodsugars are heading too low). But if youmakelittleor
no insulin, youwill still need to inject the correct dosage prior to the
snackto cover it, andto checkyourblood sugar levels 5 hourslater to
make surethey do not differ fromyour target.
For these reasons, most of my patientsdo not snackon foods that
will affect blood sugars. If you do snack, the same carbohydrate limit
that applies to meals should also beapplied to snacks. If youconsume
12 grams of carbohydrate for lunch and for dinner, 12 grams of car
bohydrate would be the upper limit for carbohydrate for any single
snack. Lesser amounts of carbohydrate for a snack— asthe Laws of
Small Numbers wouldsuggest — willnaturally poselesser problems.
If you're hungryseveral hours after ameal, checkyour bloodsugar be
fore snacking. Hunger mayreflect hypoglycemia, reflecting in turn too
much insulin, and shouldbe treated with glucose tablets asindicated in
Chapter 20 anda possible reduction of insulin dosage the next day. An
important rulefor alldiabetics: When hungry, check blood sugar.
Estimating the Dose of Regular Insulin for a Snack
There areseveral different approaches to this problem.
The simplestisto decide in advance that youwill eat for your snack
exacdy halfthe amount of carbohydrate and protein that youeat for
lunch or supper. Remember that fat has no direct effect on blood
sugar, soyouneedonlyconsider thecarbohydrate andprotein. Cover
the snack with exacdy half the dose of regular that you take for the
Intensive Insulin Regimens 299
meal you selected. If your snack is one-third or one-quarter of your
selected meal, then you'd naturally take one-third or one-quarter your
usual dose of regular for this meal — rounded off to the nearest V4
unit. You should inject the regular insulin as far in advance of the
snack as you would for a meal. In a pinch, you can take lispro instead
of regular and wait 20 minutes instead of, say, 45 minutes before
snacking. But for this insulin, take only two-thirds of the dose that you
would take for regular.
If you select a snack containing carbohydrate and/or protein that is
not in the same proportion as one of your meals, use the computa
tional method outlined on pages 292-297 for regular meals. To test the
validity of your computations, skip lunch and lunch insulin and take
the snack and snack insulin instead. Check your blood sugar before
taking the snack insulin, and then check it again 5 hours after eating.
This will help you determine the dosage correction to make when you
next decide to do the same experiment (perhaps a few days later, as
you may not wish to skip lunch two days in a row). You may want to
try this several times to be sure of the dose. Thereafter, you won't have
to skip lunch in order to have a snack.
If you've decided that your snacks will consist only of a small
amount of protein (say, lessthan 3 ounces) and no carbohydrate, you
can take your regular insulin 20 minutes before eating instead of 45
minutes before. This is because protein is converted to glucose much
more slowly than is carbohydrate. Be sure to keep the protein and/or
carbohydrate content of your snack(s) the same from one day to the
next, as you probably won't want to do more experiments to deter
mine doses of insulin.
Last but not least, blood sugars will beeasierforyou tocontrol ifyou
don't snack at all, or if you makeyour snacka small amount of sugar-
free Jell-O (without maltodextrin) instead of real food. It is important
to remember thatfor covering a meal orsnack, the dose of most-rapid-
acting insulin (lispro) should always be only two-thirds the equivalent
dose of regular.
In my experience, random blood sugars of nonobese, nonpregnant,
nondiabeticadults tend to clusterclosely around 83mg/dl (4.6mmol/1).
300 Treatment
Childrentend to run slighdylower. About 1hour aftera high-carbo
hydrate meal, many nondiabetics may have considerably higher val
ues. This, however, isnot"natural," because for mostof humanhistory
priorto the development of agriculture about 10,000 years ago, high-
carbohydrate mealswerenot usually available. Americansnow eat an
average of morethan 150 pounds of added sugar peryear, something
that the average human would not have experienced in a lifetime
10,000 years ago. Nowadays fast-acting carbohydrate accounts for the
largest part of energy consumption. So if we ignore elevatedblood
sugars that may be encountered shordy after high-carbohydrate
meals,a"normal"valuewould be 83mg/dl, perhaps evenlower.
Several recent studies have demonstrated that risk for both cardiac
andall othercauses of death increases as bloodsugars or the equiva
lentvalues of HgbAlc exceed about 75 mg/dl.
With the above information in mind, for type 2 diabetics who use
no or very little injected insulin, I seekblood sugars of 80-85 mg/dl
(4.4-4.7 mmol/1). Since type Is and type 2s who inject nontrivial
amounts of insulin cannot turn off injected insulin as their blood
sugars drop, there always exists the possibility of going too low(hypo
glycemia). I therefore throwin asmallsafetyfactor and asksuch indi
viduals, at least initially, to shoot for atarget of 90 mg/dl (5 mmol/1).
As you will learn in the next section, we try to correct blood sugars
when they areaboveor belowatarget. Sincewe follow alow-carbohy
drate diet, our targetremainsthe samebefore, during, and aftermeals,
as it probablywas for our distant ancestors.
Under certaincircumstances we will set ahigher target:
• If someone's blood sugars prior to starting our regimen were
very high, she/he will experience the unpleasant symptoms of
hypoglycemia at blood sugars that are well above our 85 or 90
mg/dl. Thus if a new patient has had most blood sugars in the
vicinity of 250mg/dl, we might initially set atarget of 140 mg/dl.
We would then lower this target slowly overa periodof weeks.
• Sincethe initialcalculations of insulindoses may be too high, in
spite of the precautions describedearlier, it is wise to have a sub
stantial safety factor. Thus one might set an initial target of 120
mg/dl and then slowlylowerthis to 90 over a period ofweeks af
ter it becomes apparent that no blood sugars less than 70 mg/dl
have been encountered. This safety factor may also protect pa-
Intensive Insulin Regimens 301
tients who at first make mistakes, because it is difficult to follow
everything taught herein perfecdywhen starting out.
• Some insulin users, for whatever reason, are not meticulous in
following what they have learned in this book or in my orifice —
most commonly the dietary guidelines. These folks will in
evitably experience roller-coaster blood sugars, although to a
much lesser degree than in the past. Here again it is safer to use a
target well above 90 mg/dl. A similar problem is encountered
with people who experience unpredictable exercise, such as la
borers and small children.
• Insulin pump users experience much greater uncertainty of in
sulin absorption than do those who inject. We therefore find it
necessaryto shoot for a higher than normal blood sugar, just to
reduce the likelihood of severehypoglycemia.
• Last but not least are those with gastroparesis (see Chapter 22).
Here the unpredictable variations in blood sugars are great
enough that a higher long-term target is frequendy necessaryin
order to avoid verylowvalues.
Sooner or later a dietary indiscretion, an infection, morning exercise,
acute emotional stress,or evenerrors in estimatingmeal portions may
cause your blood sugar to rise substantiallyover your target value. If
your beta cellsare still capable of producing moderate amounts of in
sulin, your blood sugar may drop back to target within a matter of
hours. On the other hand, youmaybelikeme and makelittleor no in
sulin,or youmaybe veryresistant to your owninsulin.If anyof these
is the case,your physicianmay want you to inject lispro, glulisine, or
aspart whenever your blood sugar goestoo high.*(Asyou've probably
noted, doses of insulin used to bring down elevated blood sugars are
often referred to as"coverage.") Because thesework faster than regular
* Because lispro is somewhat faster actingthan aspart, I prefer the lispro for cor
rectingelevatedblood sugars.
302 Treatment
insulin, they are much preferred for this purpose. (If you are presently
covering elevated blood sugars with regular insulin, use care when
switching to lispro — see "Some Final Considerations Regarding
Lispro, Aspart, and Glulisine Insulins," page 312.) To do this properly,
you must first know how much Vi or 1 or 5 units of lispro insulin will
lower your blood sugar; it's usually 50 percent more than regular will.
This requires yet another experiment.
Wait until you have a blood sugar that is at least 20 mg/dl above
your target (but this should not be an elevated measurement taken on
arising — the dawn phenomenon can muddy the result of the experi
ment). To make sure that your prior mealtime bolus dose has finished
working, this blood sugar should be measured at least 5 hours after
your last dose. Besure that you havetaken your morning basal dose of
glargine or detemir. For this test, skip your next meal and the insulin
bolus that covers it.
Now refer to Table 19-1, which suggests the amount that 1 unit
of lispro might lower your blood sugar, for the purpose of this trial
only. The left-hand column represents the sum of your daily doses of
detemir/glargine or NPH that you are taking just to keep your fasting
blood sugars level (your basal doses). The middle column shows the
amount that 1unit of lispro will probably lower your blood sugar. The
right-hand column showsthe amount that 1unit, as read on a syringe,
would likely lower blood sugar using a dilution of 3:1. (See page 273
for diluting instructions.) Again, this table isonly approximate. Its only
purpose is to suggest how much lispro (H) you might try for this ex
periment. The column for diluted insulin is for those few individuals
(children, for example) who find that a little goes a long way.
After recording your elevatedblood sugar, determine the amount of
lispro insulin suggestedby the table to bring your blood sugar down to
your current target. Let's assume that the sumof the dosesof detemir/
glargine/NPH that will just keep your blood sugars level (if no meals)
is 9 units. Then, by interpolating between lines in the table, 1 unit of
lispro will probably lower your blood sugar about 54 mg/dl. Let's fur
ther assume that your blood sugar at the time of this experiment is 175
mg/dl and that your target is 100 mg/dl. You therefore would like to
lower your blood sugar 75 mg/dl. Dividing 75 by 54 yields 1.38 units
lispro. Rounding down to the nearest quarter-unit, VA units of H
should lower you about 1.25 x 54 = 68 mg/dl. This is certainly close
enough, so you would inject \lA units.
Intensive Insulin Regimens 303
TABLE 19-1
1 unit H (full
strength) might lower
blood sugar
Total daily basal dose
of undiluted long-acting
and intermediate-acting
1 unit H (3:1 dilution)
might lower
blood sugar
2 units 240 mg/dl 13.3 mmol/F 60 mg/dl 3.3 mmol/1
3 180 10 45 2.5
4 120 6.7 30 1.7
5 96 5.3 24
6 80 4.4 20
7 68 3.8 17 0.9
8 60 3.3 15 0.8
10 48 2.7 12 0.7
13 37 2.1 9 0.5
16 30 1.7 7.5 0.4
20 24 1.3 6 0.33
25 19 1.1 5 0.27
*If you have gastroparesis and must take more longer-acting insulin at bedtime in
order to cover overnight emptying of your stomach, instead of using your bedtime
dose to arrive at this number, substitute double your morning long-acting dose and
don't add in the bedtime dose.
Reminder: mmol/1, or millimolesper liter, is the standard international measure of
blood glucose level (1 mmol/1 = 18mg/dl).
Check and record your blood sugar again 4,5, and 6 hours after the
shot.* The lowest value will not only tell you how much your blood
sugar dropped but also how long it took. For most of us, we assume
that the lispro/aspart/glulisine finishes working in about 6 hours. If
your lowest value occurs at or after 6 hours, you should in theory wait
at least this long in the future before checkingyour blood sugars to see
*Note that this experiment requires that you refrain from eating for 5 hours af
ter your last shot of regular and then another 6 hours after the lispro, for a total
of 11 hours. With luck,you'll have to do this onlyonce in your life.
304 Treatment
if the extra shot of lisproreally brought you down to target. Let's say
that the 1V4 units of lispro in the above examplebrought your blood
sugar from 175 down to 81 mg/dl after 5 hours, and it did not drop
further at 6hours.Nowyou've learned that 1XA units Hwillloweryour
blood sugar by 94mg/dl (or 175 - 81). Divide 94mg/dl by 1.25 (units
of lispro/aspart/glulisine) to find that 1unit Hwill actually loweryour
blood sugar 75 mg/dl. Whatever this value turns out to be, write it
down on your Glucograf data sheet in the box l unit will
lower blood sugar. In this case, we have learned that our initial es
timate that 1unit would lower you 54mg/dl wasoff by about 39 per
cent.This canhappen, andthisis preciselywhywedo this experiment.
If at any point during this experiment your blood sugar drops 10
mg/dl or more belowyour target, immediately correct to target with
glucosetablets,asdetailedin the next chapter. This will offset the haz
ardof hypoglycemia. On your data sheet, record the number of glu
cose tablets that you used. After you have read the next chapter, you
will understandhowknowingthe number of tablets will enable you to
completethe above calculation without terminating or repeating the
As statedat the beginningof this chapter, it shouldn't be necessary
for you to perform any of the abovecalculations on your own. This is
the job of your health care professional, who can use our table and
should havemuch more experience than you. Heor she might want to
try a simple option. For example, your doctor might instruct you to
measureyour 6-hour, posdunchblood sugar and, if it'sover 180 mg/dl,
to inject 1unit oflispro(H) andseehow far your blood sugar drops in
another 6 hours (without eating again). This will tell you approxi
mately how much 1unit will loweryour blood sugar.
Once you know how much 1 unit of lispro will lower your blood
sugar, you're in a positionto bringdown yourblood sugar rapidlyif it
goesmuch aboveyour target. All you needto do is to inject the proper
dose. Within hours, your blood sugar will probablyreturn to target,
unless for you small doses of lispro work more slowly, or unless you
have a very high blood sugar and take more than 7 units (split into
smallerdoses,eachno greater than 7 units). These extradoses arewhat
is known as coverage. Once your insulin doses have been fine-tuned, it
Intensive InsulinRegimens 305
should rarelybe necessary for you to cover with more lispro than will
lower you 60 mg/dl, unless you overeat, have an infection, or suffer
from gastroparesis.
Never cover anelevated blood sugar with lispro ifyouhave notwaited
for the last dose of regular or lispro to finish working. After all, if two
doses are working at the same time, your blood sugar can drop too
low. This is one reasonyou should knowhowlong it takes for a dose
of regularor lispro to complete its action.
It is convenient to assumethat the rapid- and most-rapid-acting in
sulins continuetheiractionforonly5hoursafterinjection, eventhough
Table 17-1shows8-10 and 6-7 hours, respectively. The reason for our
not-quite-correct assumption is that if we assumed even 6 hours for
completion of action and you were to correct blood sugars at least 6
hours after eachinjectionof the rapidinsulins (e.g., beforemeals and
at bedtime), you would not only have inconveniendy spaced meals
and bedtime, but also would have to remain awake at least 18 hours
daily, leaving you only6 hours of sleep. We thereforeassumea 5-hour
action time for convenience, in the reasonable assumption that blood
sugar drop thereafter will be minimal. Thus we arrive at a prime
guideline for correcting elevated blood sugars: Wait at least 5 hours
from your last shot of a rapid-acting insulin before correcting ele
vated bloodsugars.
Suppose target blood sugar is 90 mg/dl and you wake up in the
morning and find that your fastingblood sugar is 110,an elevation of
20mg/dl. If 1unit of lisprolowers you40,you'dimmediatelyinject Vi
unit as coverage. If you plan on havingbreakfast in 40 minutes, just
takethis xh unit as a separate shot, in additionto your usualbreakfast
dose of regular.
Another timeyoumayfind that 5hoursafter yourlunchtime regu
lar was injected, your blood sugar is 60mg/dl above your target. If 1
unit of lispro lowers you40mg/dl, take Vh unitsof lispro right away.
A major variation from the 5-hour rule applies to children and to
anyone whosleeps more than 9hours overnight and has meals spaced
less than 5 hours apart. These people should correct elevated blood
sugars onlyupon arising in themorning, unless an alarmisset to ring
during the night for a 5-7 hour postdinner correction. This alarm
could also signal thetimeforthenightiy dose of longer-acting insulin
sothat less than 9hours elapses between the night doseand the morn
ingdose,as dictated by the dawn phenomenon (seebelow).
At first, after you cover with lispro, you maywant to check your
306 Treatment
blood sugar when the insulin has finished working, to make sure that
the numbers from your original experiment were correct. After a few
times, however, you will become confident that your calibration is
Many of us find that blood sugar increases during the short interval
between arising in the morningandeating breakfast. Thisiswhy, dur
ingthe first week or twoon our regimen, it isnecessary to check blood
sugar not only upon arising but also when you sit down for breakfast.
If such an increase occurs regularly, you should cover it before it oc
curs — on arising— with the appropriate dose of lispro. This will
prevent the dawn phenomenon increase. I find it necessary to take Vi
unit of lisproeverymorning upon arising, in addition to any coverage
I might need for a slightlyelevated fasting blood sugar.
Under certain circumstances, lispro insulin will not lower your blood
sugar as much as you would expect based upon your calibration. Let's
take a look at some factors that can cause this.
Your lispro is cloudy. If your blood sugar does not drop as much as
you expect, hold the insulin vial to the light to make sure that it's not
cloudy. Compare it with a fresh vial to be sure. Lispro insulin should
be crystal clear; if it is even the slightest bit cloudy it has been deacti
vated and should be discarded. Also discard the vial if it has been
frozen, kept in a hot place, or kept out of the refrigerator for more
than three months, since temperature extremes will also affect its po
tency. According to the manufacturer, glulisine is more likely to deac
tivate at elevated temperatures than lispro.
Your fasting blood sugar was high on arising in the morning
and you can't get it down. The dawn phenomenon causes more in
sulin resistance in the morning for some people than for others. If you
start the day with an elevated blood sugar, you may require more
IntensiveInsulinRegimens 307
lispro coverage to bring down the elevation at that time than you
would 4 or more hours later in the day. If you find that early-morning
lispro coverageis not very effective, reviewyour blood sugar profiles
with your physician. You'll probably be told that you should increase
your coverage by one-third, one-half, or some other proportion dur
ing the first fewhours after you wakeup. More than 3 hours after aris
ing, this increase in any coverage should no longer be necessary.
To prevent overnight blood sugar increases, be sure to wait at least
5 hours betweenyour supper premeal bolus and your bedtime basal
insulin and correct accordingly. You may do better by reducing your
protein at supper, and certainly you should eat no bedtime snacks. Be
sure to take your bedtime insulin lessthan 9 hours before your morn
ing basal dose.
Your blood sugar was higher than 200 or 300 mg/dl. At suchhigh
bloodsugars webecomemoreresistant to theeffects of injected insulin.
This increased resistance may become very significant as blood sugar
rises above250mg/dl. But the point at whichresistance becomes signif
icant isnot precise, and its magnitudeisdifficult to determine. Werarely
encounter such high blood sugars once insulin doses and diet are ap
propriate. If you do measurea veryhighbloodsugar, coverit with your
usual caUbration for lisproand wait the usual 5 hours or so. Then check
your blood sugar again. If it has not comeall the waydown to your tar
get, take another coverage dose based on the new less elevated blood
sugar. This time the coverage willprobablybe fully effective.
Infections. If your lispro coverageor any other insulin dose is lessef
fective than usual, you may havean infection. Weonce discoveredthat
a patient had an intestinal inflammation calleddiverticulitis only be
causehe waswiseenough to telephonemewhenhis blood sugarswere
a litde lessresponsive to insulin than usual. It's important that youno
tifyyourphysician whenever youfind thatyour insulin appears to belos
ingitsefficacy. SeeChapter 21.
Intramuscular shots of insulin can be quite useful for bringing down
elevatedblood sugar more rapidly than our usual subcutaneous shots.
308 Treatment
You should not ordinarily use them for your usual meal doses of reg
ular insulin, and you should never inject glargine, detemir, or NPH
into a muscle— it makes no senseto speed up the action of a long-
acting insulin.
Typically, an intramuscular shot of lispro will begin to lower an el
evatedblood sugar within about 5 minutes. It will finish acting about
1hour sooner than your usual subcutaneous injection, and it will have
your blood sugar closeto your target within about 3 hours.
Problems to Consider
With intramuscular shots, you may experience several problems that
you do not encounter with subcutaneous shots. Because of this, I
give my patients the option of usingor not using this method of self-
injection, and fully appreciate the feelings of those few who turn it
down. Here are some obstacles you mayconfront:
Fat arms. If you havefat arms, don't eventry intramuscular shots. If
you havea lot of fat over the deltoidmuscleon your upper arm (Fig
ure 19-2), the needleon your insulinsyringewillbe too short to pen
etrate the underlyingmuscle.
Missing the muscle. Even moderatelyslimpeoplesometimes"miss"
the muscle because eventhe longer (Vi-inch) needle sometimes may
not penetrate deeplyenough. Sincewe cannot always tell whether or
not the needle hit the muscle, all of us must wait as long before
recheckingour blood sugars as wewould for a subcutaneous shot.
Hitting a blood vessel. You aremuch more likely to hit a blood ves
sel than with subcutaneous injections.This can be brieflypainful. You
can alsoget blood on your shirt if you shoot right through the sleeve
as I do. I estimate that I hit a blood vessel once in everythirty intra
muscular injections. (See page 261 for instruction on using hydrogen
peroxide to remove bloodstains from clothing.)
Pain. If for whatever reason you're unable to throw the needle in
rapidly like a dart, all your intramuscular shots may be briefly
painful — if so, do not evenbother to attempt them.
IntensiveInsulinRegimens 309
Intramuscular Injection Technique
Please refer to Figure 19-2 as you read the following step-by-step in
structions. Do not use a syringe with the new short needles. I keep on
hand a supply of syringes with Vi-'mch needles just for intramuscular
1. Locate your deltoid muscle, illustrated in Figure 19-2. It begins at
the shoulder and ends about one-third of the waydown your up
per arm. It's wide at the shoulder and tapers to a Vshape farther
down. You may be able to feel the Vwith your fingers if you lift
your arm to the side until it is parallel to the floor. This will
tighten the muscle and make it feel harder. We usually use the
deltoid muscle because it is easy to find, is relatively large and
thick, and is less likely to be covered with a deep fat pad than
most other muscles.
2. Now, allowyour nondominant arm (left if you're right-handed)
to dangle loosely at your side. This will relax the muscle, so that
the needle can penetrate easily.
3. The site for injection will be near the upper (wider) end of the
deltoid, about \l/z inches below your shoulder (at about the po
sition of the arrow in Figure 19-2). We use the wide end of the
muscle because you are less likelyto miss it with the needle, and
because you would not want to pierce the axillarynerve, which is
located near the tip of the V, at the lower end.
4. As your nondominant, target arm dangles loosely at your side,
pick up the syringe with your dominant hand and "throw" the
Terry Eppridge
Fig. 19-2. Thedeltoidmuscle (arrow
indicates preferred sitefor intramus
310 Treatment
needle straight into the injection site as you would a dart — but,
of course, don't let go of the syringe. Do not grab any flesh, as
you do for subcutaneous shots. Do not inject at an angle, but go
in perpendicular to the skin. Befast,as a slowintramuscular shot
can hurt. Push in the plunger rapidly to inject your insulin. Now
pull out the needle. Touchthe injection site with your finger, to
make sure you have not bled.
5. If the shot hurts, you probablyhit a small blood vessel, so be pre
pared for some blood. In such a case, press the injection site
firmlywith a finger. Hold it there for about a minute. This will
prevent or stop any bleeding. If you do not press, you will de
velopa slightiypainful lump wherethe blood accumulatesunder
the skin. The lump will turn yellow or black and blue after a
number of hours. If you inject through your shirt or blouse and
get it bloody, applyhydrogenperoxide,as described on page 261.
Once you havegiven a number of intramuscular shots using your
dominant hand to operate the syringe,try switchinghands and arms.
This mayseemcumbersome at first, but withpractice youwillbe able
to inject into either arm.
As indicated previously, if you are still making some insulin and are
insulin-resistant, you maybe ableto take rosiglitazone instead of reg
ular insulin before certain meals. This will depend upon your post
prandial blood sugar profile. There may be no therapeutic advantage
to such a substitution, but it might be more convenient. Remember,
however, that you will probablyhaveto wait at least 60minutes before
starting your meal. It is usuallymore convenientto take a shot of reg
ular insulin, since the waiting time after injecting is generallyonly 45
A more important use for an ISA in combination with insulin oc
curs if you are overweight or have polycystic ovarian syndrome
(PCOS;seeAppendixE) and your bedtime dose of long-acting insulin
is more than 8-10 units. This suggests that you may have insulin resis
tance, which may respond to one of the ISAs or insulin mimetics. Re
callthat ISAs increaseyour sensitivityto insulin. Large dosesof insulin
help build fat, of course,and can alsocausefurther down-regulation,
Intensive Insulin Regimens 311
or desensitization, of insulin receptors. If you'reobese,the lessinsulin
you have in your system storing awayfat, the better. So there may be
some advantageto reducingyour bedtime insulin dose.
If your physician decides to add one ofthese agents to your bedtime
regimen, he or she will want you to build up the dose gradually while
simultaneously reducing your dose of basal insulin. She/he might
want to beginwith the extended-release versionof metformin because
it will keep working all night and is not likelyto cause the digestive
discomfort sometimes found with the more-rapid-acting version. If
your bedtime insulin requirements are not reduced while taking the
maximum recommended doses of severaloral agents before sleeping,
then the bedtime oral agents are serving no purpose and should be
discontinued. The FDA warns against using rosiglitazone or pioglita-
zone for people taking insulin, as there is a small risk of congestive
heart failure (due to fluid retention) in susceptible individuals who
takeinsulin plus thesemedications.* This restrictiondoes not applyto
metformin or to the insulin-mimetic agents.
Type 1 diabetics, and those type 2s whose beta cellsare producing lit
tle or no insulin, both tend to show significant blood sugar changes
following relativelysmall changes in what they eat, their activity level,
and so on. If your blood sugarscommonlyshowchangesof more than
20 mg/dl in the course of a day, you probably should measure blood
sugar profiles dailyfor the rest of your life. Suchfrequent monitoring
is necessaryso that you can correct high blood sugars with lispro or
lowblood sugars with glucose tablets (seenext chapter).
I've seen many individuals on our regimen whose blood sugars are
quite stable even though they require the 5 daily shots typical of in
tensive insulin therapy.Thesepeopleusuallyrequire small dosesof in
sulin, typically for adults under 8 units dailyfor all dosescombined. If
*Insulinin largedosescauses fluidretention.Rosiglitazone and pioglitazone can
alsocause fluid retention. It is likelythat the people reported to have developed
heart failure while taking both rosiglitazone or pioglitazone and insulin were
takinglargedoses of insulin to coverthe usual high-carbohydrate diabetes diet.
312 Treatment
you fit into this category, your beta cells are probably still producing
some insulin. This enables your system automatically to smooth out
the peaks and valleys that your blood glucoseprofile would otherwise
show. With such stable blood sugars (varying less than 20 mg/dl
daily), there's no reason to bother taking dailyblood sugar profiles.
You would, instead, prepare a full blood glucoseprofile (fiveto seven
tests) for 1 day every2 weeks. If you spotted a change in your blood
sugar ranges,you'd checkthe next few daysto see if it continued. If it
did, you would contact your physician, who might want to explorethe
possiblereasons for such changes. If youbecome ill, or if,say, you have
a school-age child who brings home a cold, you might want to check
your blood sugar profiles everyday. If your physician has prescribed
oral or injected steroids for other disorders such as asthma or bursitis,
you should be checking and recording blood sugars, as they will cer
tainly increase.
Perhaps as a result of readingone of myprior books, you may already
be covering elevated blood sugars with regular insulin. If this is the
case, be very careful when using lispro, glulisine, or aspart for this
purpose. I and many of my patients have found them to be more ef
fective than regular — that is, a given dose is likely to lower blood
sugar more than the same dose of regular. For example, I find that
while 1 unit Rwill lowermyblood sugar 40 mg/dl, 1unit H will lower
it 60 mg/dl. I advise, therefore, that you initially take two-thirds as
much lispro (H) as your prior regular (R) for this purpose. Based
upon the initial effect on your blood sugar, you can then adjust subse
quent dosesof theseanaloginsulins.The same consideration applies if
you eat out and use H to covera meal.
We have also observed that when lispro is used to cover meals,
blood sugars are lesspredictablethan with regular.This result was not
mentioned in reports of clinical trials of lispro, probably because the
trial population followed a high-carbohydrate diet and had such wide
blood sugar fluctuation that this effect was not apparent. In spite of
this consideration, I believe that lispro greatiybenefits those who use
it properly.I am most grateful for its availability.
IntensiveInsulin Regimens 313
It's certainly worth mentioning that lispro is available in small
(3 cc) cartridges. These can becarried inajacket pocket or small purse
without creating an unsightly bulge or the need for refrigeration.
When using cartridges, insert aneedle andpull back ontheplunger of
the syringe very slowly. Do not inject airinto these cartridges. If you
draw out too much insulin, do not inject it back into the cartridge.
Squirt the excess into a plant or wastebasket, et cetera.
Aspart and glulisine function on a par withlispro, although some
what lessrapidly.
Mucheffortand expense arebeing devoted to promoteand marketin
sulin pumps. These devices were designed to make multiple daily in
jections easier. Theyalso doawaywiththeneed forlong-acting insufins.
The instruments consist of two basic elements:
• Apump unit about the size of a small pocket calculator, which
you can hang from abelt, keep inapocket, or pintoyourclothing.
• Large-bore plastic tubing that stays in your skin, typically just
above your waist. The plastic tubing, which is inserted into the
skin through a large, retractable needle, usually should be
changed every2-3 days.
The pump unit can beloaded with a supply of lispro or other very
rapid acting insulin that lasts a number of days before refilling is nec
essary. It delivers a tinybasal flow of insulin all daylong, giving an ef
fect similar to that of 2 daily injections of long-acting insulin. This
basal rate canbepreset bytheuser andcan even besettochange auto
matically at various timesof the day. Premeal bolusor corrective doses
arereadily produced bysetting thedose andthenpushing a button.
Insulin pumps offer the following advantages over multiple daily
• There is no need to carry a number of insulin syringes when
away from home, but catheters and other supplies must accom
pany you.
• Correctiveinjections are elegantiy simple.
• Pumps canbesettoautomaticallyincrease thebasal delivery rate
shortly before arising in the morning, thereby circumventing
314 Treatment
problems associated with the dawn phenomenon. They thus ren
der it unnecessaryfor you to arise earlyon weekends to take your
long-acting insulin.
On the other hand, insulin pumps can pose some problems:
• Pump failure, tubing comingout ofthe skin, insulin coagulation,
tubing blockage, or kinking can occur in spite of sophisticated
alarms and safeguards. As a result, ketoacidosis has occurred
overnight in many type 1users.
• There is a moderate incidence of infections at injection sites.
Manyofthesehaveformedabscesses requiring surgical drainage.
• Severe hypoglycemia is more common among pump users, pos
siblybecause of mechanical problems.
• Insulin pumps cannot be used to give intramuscular injections
for more rapidlowering of elevated blood sugars.
• All of the long-term (seven-plus years) pump users that I have
seenhad fibrosis (scar tissue formation) at the injection site.This
had impaired their insulin absorption so much that even high
doses failed to control their blood sugars. In addition, blood
sugar effects of pump boluses appeared to be inconsistent in
these individuals.
• Until recendy, pump delivery rates could not be set for less than
0.1 units per hour. This makes it necessary forbasal dosing to be
in multiples of 2.4 units per day, thereby preventing fine adjust
ments of basal insulin. For example, I require3 units of glargine
twice daily formybasal insulin.With apump I would haveto take
either too little — 4.8 units (2 x 2.4) — or too much — 7.2 units
(3x 2.4).Some pump manufacturers nowmaketheir product ad
justable to 0.01 units perhour, which overcomesthis problem.
• Many people are turned off by the idea of constandy having
large-bore tubing stickingin their abdomens.
• Usersexperienceat least some inconvenience with the four S's—
sleep, showers,swimming, and sex.
In our experience, insulin pumps do not providebetter blood sugar
control than multiple injections. Contrary to a common misconcep
tion, they do not measurewhat your blood sugaris and correctit auto
matically. Furthermore, most pumps areprogrammedto producemeal
Intensive Insulin Regimens 315
boluses that are computedto cover varying amounts of carbohydrate,
totallyignoring both dietary protein and the Laws of Small Numbers.
Recendy an insulin pump (OmniPod) has become available that
usesboth a slimmer needle (28gauge) and a short lengthof veryfine
tubing. The pain is virtuallyeliminated, and the long-termproblems
causedbya largeforeign body(i.e., the tubing) under the skinare con
siderablyreduced. The basal infusion rate, however, is still too great
(0.5units per hour) for most peopletakingphysiologic doses of basal
insulin. This may be improved in the future.
In 2006 the FDA approved a powdered human insulin inhalable
through the mouth and absorbed in the lungs. It is manufactured by
Pfizer and carries the brand name Exubera. Similar products are un
der development by other manufacturers.
So at long last, insulin can be administered without puncturing the
skin with a needle. Many physicians and members of the press are
praising this accomplishment. Is it reallya benefit to diabetics, or is it
too good to be true?
Exubera is dispensed in 1mg and 3 mg packetsand aerosolizedfor
inhaling by a plastic "puffer" measuring about 2 inches in diameter,
6Vi inches long when closed, and 11 inches long when open for use.
The product has been approved for use before meals by both type 1
and type 2 diabetics.
If you'vetried our painlessmethod for injectinginsulin, you proba
bly will see no advantage to puffingyour insulin. But what about the
majority of insulin users, who are takingthe usual large doses and in
jecting using the conventional "no pain, no gain" method? For those
folks, this product may be viewedas a blessing. Likewise for those who
have heard horror stories about painful injections (from no less emi
nent an organization than the Juvenile DiabetesResearchFoundation).
Manyof these peoplewill refuseto inject insulinand would rather have
high blood sugars or use the potent OHAs that eventually burn out
beta cells. Clearlysuch peoplewillmake up a major market for inhaled
insulin, so the profit potential for this product will be considerable.
Are there any likelydisadvantages to inhaling insulin? Indeed there
are. They are listed belowin random order of importance:
316 Treatment
• There is a possibility of long-term adverse effects upon lung
• The cost per dosewill be much greater than that of injectedin
sulins. It is unknown whether this cost will be coveredby many
• It will be necessary to carryaround a bulky object for premeal
• Each milligramof powderisequivalent to 1-3 units of insulin af
ter absorption. Thus people who require doses smaller than 1
unit cannot be servedby this product.
• The actual dose of absorbedinsulin can vary from one puff to
another, by 1-2 units for the 1mg packet and by asmuch as 3-9
units forthe 3 mg packet. This makes precise dosingimpossible.
This may be of uncertain importance for those eating large
amounts of carbohydrateand taking large premeal boluses, since
many diabetics are not following the methods we prescribe and
their blood sugars are on aroller coaster anyway. If you are read
ingthis book, the chances are that you are seeking to avoidor get
off the roller coaster and therefore will want to avoid the hazards
of wide blood sugar swings.
• The FDA advises against using this product if you have a cold,
smoke, or have any lung disorder such as asthma, seasonal aller
gic cough, et cetera.
Are there any diabetics who might benefit from inhaled insulin? I
speculatethat there may be many obese type 2s making considerable
insulin of their own who can't control their carbohydrate craving,
don't want to use oral agents that push beta cells to make more in
sulin, and refuse to take injectionsbut want to bring their very high
postprandialblood sugars down to levels that arelesshigh. Inhaledin
sulin may help them accomplish this. These people will likely never
have normal blood sugars, but they may not care.
How to Prevent and Correct
Low Blood Sugars
Useof medications suchasinsulin or the obsolete sulfonylurea-
typeand newer, similaroral hypoglycemic agents(OHAs) that
provoke increased insulin production exposes you to the
ever-present possibilitythat your blood sugars may drop belowyour
target value.* Because your brain requiresglucose in order to function
properly, a deficit of glucose— or hypoglycemia — can lead to some
occasionallybizarre mental symptoms. In extreme cases, it can result
in death. Although severe hypoglycemia can be dangerous, it is pre
ventable, and treatable. I encourage you to have your family, close
friends, or workmates read this chapter so they will be able to assist
you in the event you have a hypoglycemic episode and cannot correct
it alone. I mention OHAs repeatedly in this chapter because of the
hazard of hypoglycemia that they pose. Please remember that I rec
ommend insulin-sensitizing agents and insulin mimetics, while I op
pose the use of OHAs.
For our purposes in this chapter, wewill use the term "hypoglycemia"
to designate anybloodsugarthat's more than 10mg/dl belowtarget.
"Mild"hypoglycemia isanybloodsugarthat's 10-20mg/dl belowtar-
*It has beenclaimedthat the insulin-sensitizing agents(ISAs) cannot causeab
normallylowblood sugars. This is not so. As we discussed in Chapter 15, I've
seen it happen— in a very mild diabetic who was using it to facilitate weight
loss. Nevertheless, this is a rare occurrence.
318 Treatment
get. As it drops lower, it'sprogressively more "severe," and can, if left
uncorrected, becomethe condition known as neuroglycopenia, which
means"too littleglucose in the brain."
Glucose diffuses in and out of your brain slowly, whereas blood
sugar intherest ofyour bodycan rapidlydroptozero inanhour from
an intramuscular overdose of lispro insulin. Many diabetics develop
physical symptoms or signals that enable them to recognize a hypo
glycemic episode and think clearly enough to measure blood sugar
and correct it.
When blood sugar drops slowly, neuroglycopenia can occur at
about the same time that physical symptoms appear.You may not be
awareof them, however, because your brain, severely deprivedof glu
cose, is less capable of comprehending these things. "Hypoglycemia
unawareness" (reduced or absent ability to experience early signs of
hypoglycemia) is also common in individuals who have recendy had
frequent hypoglycemic episodes, because of a phenomenon called
down-regulation of adrenergicreceptors(seepage341). It can alsobe
caused by a class of cardiacdrugs (beta blockers) that slowthe heart
and lower blood pressure. If you do not notice physical symptoms,
you maythen not be able to thinkclearly enoughto realize that your
blood sugar is too low,and your cognitive state will deteriorate.
Progression of Symptoms of Neuroglycopenia
Below is a partial fist of the signsand symptoms of hypoglycemia as
they progress, ranging from mild (early) to severe (late), which to
gether make up neuroglycopenia:
• Delayed reactiontime—e.g., failure to slowdown fast enough
when driving a car.
• Irritable, stubborn behavior and lackof awareness of the physi
cal symptoms of hypoglycemia(seebox, page 322).
• Confusion, clumsiness, difficulty speaking,weakness.*
• Somnolence (sleepiness) or unresponsiveness.
• Lossof consciousness (veryrare if you do not take insulin).
• Convulsions(extremely rare if you do not take insulin).
• Death (extremelyrare if you do not take insulin).
*One study has shown these symptoms to occur whenblood sugar drops to
45-65 mg/dl.Furthermore, symptoms were foundtocontinuefor45minutesaf
ter blood sugarswerenormalized.
HowtoPrevent andCorrect LowBloodSugars 319
Some Common Causes of Hypoglycemia
In various chapters, particularly those covering insulin, we've dis
cussed a number of different potential causes of lowbloodsugar. Be
low is a list of some common causes.
• Not waiting at least 5 hours after mealtime insulin before cor
recting an elevated blood sugar. This is especially dangerous at
• Too muchdelaybefore eating ameal aftertakingregular or lispro
insulinor classic OHAs, suchastheoldsulfonylureas andsimilar
newer agents.
• Delayed stomach-emptying aftera meal (see Chapter22).
• Reduced activity of counterregulatory hormones during certain
phasesof the menstrual cycle.
• Sudden termination of insulin resistance after abatement of ill
ness or stress that required higher than usual doses of classic
OHAs or insulin.
• Injectingfrom a fresh vial of insulin after having used progres
sively higher doses of insulin that hasslowly lost its activity over
a period of months.
• Switching from an insulin pump to manually injected insulin
without loweringthe dose.
• Incorrecdyassumingthat lisproisequivalent in potencyto regu
lar insulinor that biosynthetic human insulinshavethe samepo
tency as animal insulins.*
• Eating less than the plannedamount of carbohydrate or protein
for a meal or snack.
• Takingtoo much insulin or OHA.
• Engagingin unplanned physical activityor failingto cover phys
icalactivity with appropriatecarbohydrates.
• Drinkingtoo muchalcohol, especially prior to or duringa meal.
• Failure to shakevialsof NPHinsulinvigorously beforeusing.
• Inadvertentiyinjectinglong-actingor premeal bolus insulin into
a muscle.
• Injecting near a muscle that will be strenuously exercised.
*These new ultra-rapid-acting insulins have about 50 percent more blood
sugar-lowering effect than regular insulin, despite statements to the contraryby
their manufacturers.
320 Treatment
• Usinginsulin that contains protamine (NPH; seepage 264).
• Taking aspirinin largedoses, or anticoagulants, barbiturates, an
tihistamines, or certain other pharmaceuticals that may lower
blood sugar or inhibit glucose production by the liver (see Ap
pendix C).
• Asudden change from cool weather to warm weather.
Common Signs and Symptoms of Hypoglycemia
Hunger. This is the most common early symptom. A truly well-
controlled, well-nourished diabetic should not be unduly hungry —
unless he's hypoglycemic. This symptom, although frequendy ig
nored, should not be. On the other hand, hunger is also very often a
sign of tension or anxiety. One cannot assume that it automatically
signals hypoglycemia. Perhaps half of so-called insulin reactions may
merelyreflecthunger pangsprovokedbymealtime,emotional factors,
or evenhigh bloodsugars. Whenbloodsugars arehigh, the cells of the
body are actuallybeingdeprived of glucose, and you mayfeel hungry.
Thus, hunger is very common in poorly controlled diabetics. Ifyou
feel hungry, measure your bloodsugar!
Impaired visual acuity. Even mild hypoglycemia can make for dif
ficultyin readingstreet signsor fine print. More severe hypoglycemia
can cause double vision.
Elevated pulse rate. Always carry a watch with a sweep second
hand. Knowyour maximumrestingpulserate. When possiblesymp
toms of hypoglycemia appear andyouhave no handymeansof testing
your blood sugar (a sign of gross negligence), measure your resting
pulse. Manypeoplefind it more convenient to measurethe temporal
pulse (at the temple, on the side of the head between eyebrow and
hairline) or carotid pulse (on the side of the neck just belowlower
edge of jawand about 1-3 inches forward of the ear) than the radial,
or wrist, pulse. If restingpulseexceeds your maximum restingvalueby
more than one-third, assume hypoglycemia. This measurement may
be normallyelevated if you've beenwalking about during the prior 10
minutes. Yourhealth careprofessional can help you learn how to mea
sure your pulse. This exercise should never be necessary since, of
course, you haveyour blood sugar meter with you at all times.
HowtoPreventand Correct LowBloodSugars 321
Nystagmus. This symptommaybe demonstratedbyslowly moving
your eyes from side to side while keeping your head immobile. If an
other personis asked to watchyour eyes, shewillnotice—whenyour
blood sugar is low—that they may jerk briefly in the reverse direc
tion, or "ratchet," insteadof moving smoothly. You canobserve the ef
fect of this bylooking at the sweep second hand of your watch. If it
seems occasionally to jump ahead, you are experiencing nystagmus
(actually, asyour eyes jumpedto thesidefor brief instants,youmissed
seeing bits of motion of the second hand).
Absence of erections. For a man, a fairly reliable sign of early-
morning hypoglycemia is awakening without an erection, assuming
that he ordinarily experiences morning erections. Failure to experi
ence an erection when sexually stimulated likewise suggests hypo
glycemia if this is not a usual problem.
Denial. As hypoglycemia becomes moresevere, or if blood sugarhas
been dropping slowly, many patients will be certain that their blood
sugars arefine. Anobserver suspecting hypoglycemia shouldinsist on
a bloodsugar measurementbefore accepting the diabetic's denial.
Historically, the advice for correction of lowblood sugar has been to
consume moderately sweet foods or fluids, such as candybars, fruits,
cookies, hard candies, peanut butter crackers, orange juice, milk, and
soda pop. Such treatment has never worked properly, for reasons you
canprobablyguess, knowing whatyounowknowabout variousfoods
and howthey affect your blood sugar.
These moderately sweet foods contain mixtures of slow- and rapid-
acting carbohydrates. If,forexample, youeator drinkenoughthat the
rapid-acting carbohydrate in these foods raises yourbloodsugarfrom
40mg/dlup to your targetof 90mg/dloverthecourse of halfan hour,
you mayhavesimultaneously consumed so much slow-acting carbo
hydrate that your blood sugarwill go up by 300 mg/dl several hours
In the olddays, beforeI learnedto maintainmybloodsugarin nor-
322 Treatment
mal ranges, my physicians insisted that very high blood sugars after
hypoglycemic episodes were due to an"inevitable" hypothetical effect
they called rebound, or the Somogyi phenomenon.*Once I learnedto
avoidthe usual foods fortreatinglowblood sugar, I never experienced
blood sugar rebound. Nevertheless, the scientific literature does de
scribeoccasional mild insulin resistance that lasts up to 8 hours fol-
Signs andsymptomsof hypoglycemia include the following:
• Confusion (e.g., inabilityto read the time or to find
• Headache
• Hand tremors
• Tinglingsensation in fingers or tongue
• Buzzing in ears
• Elevated pulserate
• Dilatedpupils
• Great hunger
• Tight feeling in throat or near rear of tongue
• Numbness or strangesensations in lips or tongue
• Clumsiness
• Impaired abilityto detect sweet tastes
• Stubbornness
• Inappropriate laughter or joking
• Irritability
• Nastiness
• Anxiety or panic
• Pounding hands on tablesand wallsor kicking the floor
or other objects
• Miscellaneous visual impairments, such asblurredor
double vision, seeing spots, visualhallucinations (e.g.,
letters or numbers seem to be printed in Chinese)
*If your physician stillbelieves what helearnedin medical schoolabout this fic
tional phenomenon, ask him to read "The Somogyi Phenomenon— Sacred
Cowor Bull?," ArchIntern Med 1984; 144:781-787.
How toPrevent andCorrect LowBlood Sugars 323
lowing anepisode of verylowblood sugar. This isnot thedramatic re
bound caused byeating the wrong thing tobring upblood sugar.
Hypoglycemia can be hazardous, as the list of its progression on
page 318 demonstrates. We therefore want to correct it as rapidly as
possible. Complex carbohydrate, fructose, lactose (in milk), and even
sucrose, which is used in most candies — all must be digested or
• Poor physical coordination (e.g., bumpinginto walls
and droppingthings)
• Tiredness
• Weakness
• Sudden awakening fromsleep
• Shouting while asleep (or awake)
• Rapid shallowbreathing
• Nervousness
• Light-headedness
• Faintness
• Hot feeling
• Cold or clammy skin, especially on the neck
• Resdessness
• Insomnia
• Nightmares
• Pale complexion
• Nausea
• Slurredspeech
• Nystagmus (page 321)
Several ofthesesymptoms mayoccurat the sametime. One
symptom alone may be the only indicator. In some cases, there
maybe no clearly apparent early signs or symptoms at all.
324 Treatment
processedbythe liver before theywill fully affect bloodsugar. This de
laymakes these types of carbohydrate poor choices for treating hypo
glycemia. Furthermore, you need to know exacdy how much your
blood sugar will rise after eating or drinking something to raise it.
With most of the traditional treatments you must continually check
your blood sugar manyhours later to gauge the unpredictableeffect.
Raising Blood Sugars Predictably
What, then, can we useto raiseblood sugars rapidlywith a predictable
outcome? The answer, of course, is glucose.
Glucose, the sugar of blood sugar, does not have to be digestedor
converted by the liver into anythingelse. Unlike other sweets, it's ab
sorbed into the blood directiythrough the mucous membranes of the
mouth, stomach, and gut. Furthermore, aswediscussed in Chapter 14,
"UsingExercise to EnhanceInsulin Sensitivity," we can compute pre
ciselyhow much a fixed amount of glucosewill raise blood sugar. If
you havetype 2 diabetesand weigh about 140pounds, 1gram of pure
glucose will raise your blood sugar about 5 mg/dl — provided that
your blood sugar is belowthe point at which your pancreas starts to
make insulin to bring it down. If you weigh 140 pounds and have
type 1 diabetes, 1 gram of glucose will raise your blood sugar about
5 mg/dl no matter what your blood sugar maybe, becauseyou cannot
produce any insulin to offset the glucose. If you weigh twice that, or
280 pounds, 1 gram will raise your blood sugar only half as much. A
70-pound diabetic child, on the other hand, will experience double
the blood sugar increase, or 10mg/dl per gram of glucose consumed.
Thus, the effect of ingested glucoseon blood sugar is inverselyrelated
to your weight. Table 20-1 gives you the approximate effect of 1gram
glucose upon lowbloodsugarfor variousbodyweights.
If you havehandled glucose tablets, be sure to wash your hands be
fore recheckingyour blood sugar. If a source of water is not available,
lickthe fingeryou intend to prickto remove anyresidual glucose. You
can dry the finger bywiping it on your clothing or a handkerchief.
Do not keep glucose tablets near your blood sugar meter or test
Many countries have available as candies or confections products
that contain virtually all of their nutritive ingredients as glucose.
These glucose tablets are usually sold in pharmacies. Some countries
even have glucose tablets marketed specifically for the treatment of
HowtoPrevent andCorrect LowBloodSugars 325
TABLE 20-1
Body weight
1 gram glucose will raise
low blood sugar
35 pounds 16kilograms 20 mg/dl 1.11 mmol/1
70 32 10 0.56
105 48 7 0.39
140 64
5 0.28
175 80 4 0.22
210 95 3.3 0.18
245 111 3 0.17
280 128 2.5 0.14
315 143 2.2 0.12
hypoglycemia in diabetics. Table 20-2 lists a few of the products with
which we are familiar.
Of the glucose tablets listed, I personally prefer Dextrotabs because
they're veryeasy to chew, raise blood sugar quite rapidly, taste good,
are convenientiy packaged, and are inexpensive. They are also small
enough that they usuallyneed not be broken in halves or quarters to
make small blood sugar adjustments (except for children). Each jar of
100 Dextrotabs* comes with a small plastic envelope that holds 20
tablets flat. This envelope fits easily intoyour pocket or purse andcan
be refilled as often as needed. For smaller children I prefer Smarties*
or Winkies because of their tiny size. Most glucose tablets begin to
raise blood sugar in about 3 minutes and finish after about 45 min
utes, if youdon't have gastroparesis (if youdo, see Chapter 22).
With thisbackground in mind,howshould youproceed whenyou
encounteralowblood sugar?
*Available at Rosedale Pharmacy, (888) 796-3348.
TABLE 20-2
1 tablet will raise
Country of
Name of
Grams of
per tablet
blood sugar of
140-pound person
with low blood sugar
USA Dextrotabs 1.6 8 mg/dl 0.44 mmol/1
USA Sweetarts or
2* 10 0.56
USA, UK, B-D Glucose 5 25 1.40
Canada Tablets
USA, Canada Smarties or
0.4 2 0.11
USA, Canada Dex4 4 20 1.10
UK, Canada Dextro Energy 3 15 0.83
FRG Dextro-Energen 4 20 1.10
Tablet size may vary.
'Availableat kosher stores and at Rosedale Pharmacyor at www.smartiesstore.com.
Ideallysuited for childrenbecause of their smallsize.
If you experience any of the symptoms of hypoglycemia detailed
earlier — especially hunger — measureblood sugar. If blood sugar is
10 mg/dl or more belowtarget, chewenough glucosetablets to bring
blood sugar back to your target. If you have no symptoms but dis
covera lowblood sugar upon routine testing, again,take enough glu
cose tablets to bring blood sugar back to your target. Having no
symptoms is not a valid reason for not taking tablets. A low blood
sugar without symptoms carries more risk than one with symptoms.
If you weighabout 140pounds and your blood sugar is 60 mg/dl but
your target is 90 mg/dl, then you might eat 4 Dextrotabs. This would
raiseyour blood sugar, accordingto Table20-2, by 32 mg/dl, bringing
you to 92 mg/dl. If you are using Dextro-Energen, you'd take 1V4
tablets. With B-Dtablets, you'd take 1. Simple.
HowtoPrevent andCorrect LowBloodSugars 327
If your lowbloodsugar resulted from taking too much insulinor
OHA, it may continue to drop after taking glucose if the insulin or
OHA hasn't finished working. You should therefore recheck your
blood sugar about 45minutes aftertakingthe tablets, to rule out this
possibilityandto see ifyou're backwhere youbelong. Ifbloodsugaris
still low, take additional tablets. Ifyou have delayed stomach-emptying,
you may have to wait as much as 2 or more hours for full effect.*
What if you'reout of your homeor workplace and don't have your
blood sugar meter? (A major crime, as noted earlier.) If you think
you'rehypoglycemic, playit safeand takeenoughtablets to raiseyour
blood sugar about 60 mg/dl (7 Dextrotabs, for example, or 2 B-D
tablets). You mayworrythat this will bringyou too high. If you take
insulin, this poses no problem. Simply checkyour blood sugar when
yougetbackto your meter. If it'sabove your target, takeenoughlispro
(or aspart or glulisine) to bring youbackto target, but be sure to wait
5 hours after your last dose of rapid-acting insulin. If you don't take
insulin, your blood sugar should eventuallycome back on its own, be
cause your pancreas is still making some insulin. It may take several
hours, or even a day, depending upon how rapidly you can produce
insulin. In anyevent,you mayhavesaved yourselfan embarrassingor
even disastrous situation.
Take your glucose tablets anyway. If you don't, you may become very
hungry, overeat, and be too high many hours later. The medication
you take for a meal is intended to keep your blood sugar level. Soif it
wastoo lowbeforea meal, it willbe too lowafter if you don't takeyour
glucose but eat properly.
*This time frame can be gready reduced by drinking a glucose solution (see
page 377).
328 Treatment
Manyofthesymptoms ofhypoglycemia are actuallyeffects of thehor
moneepinephrine (which youmay knowasadrenaline). If youdo not
have the problems listed in the section "Hypoglycemia Unawareness"
on page 340, your adrenal glands will respond to hypoglycemia by
producingepinephrine. Epinephrine, like glucagon, signals theliver to
convert stored glycogen to glucose. It is epinephrinethat brings about
such symptoms as rapid heart rate, tremors, pallor, and so on. (Beta
blocker medications can interfere with the ability of epinephrine to
cause these symptoms.) Epinephrine has a half-life in the blood of
about 1 hour. This means that an hour after your blood sugar comes
back to target, about half the epinephrine you made is still in the
bloodstream. This can cause a persistenceof symptoms, even if your
blood sugar is normal. Thus, if you took some glucose tablets an hour
agoand still feel symptomatic, check your bloodsugaragain. If it's on
target,try to control thetemptationto eat more.If your bloodsugaris
still low, more tablets are warranted.
Mild to moderate hypoglycemia can cause severehunger and an asso
ciated panic. The drive to eat or drink large amounts of sweet foods
can be almost uncontrollable. Newpatients, before starting our regi
men, havetold me storiesof eatingan entire pie, a jar of peanut but
ter, a quart of ice cream, or drinking a quart of orange juice in
response to hypoglycemia. Before I stumbled onto blood sugar self-
monitoring and learnedhow to use glucose tablets, I did much the
same. The eventual outcome, of course, was extremely high blood
sugar several hours later.
Since the effects of glucose tabletsare so predictable, the panic ele
ment has vanished for me and for most of my patients.
Unfortunately, rapid correction of blood sugar does not always
correct the hunger.This maybe somehowrelatedto the long half-life
of epinephrine and the persistence of symptomsevenafter restoration
of normal blood sugars. My patients and I have successfully coped
HowtoPrevent andCorrect LowBlood Sugars 329
with this problem in avery simple fashion. You can trythesame trick
we use.
First, consume the appropriate number of glucose tablets.
If overwhelming hunger persists, consider what might satisfy it.
Typical options include afull meal (such as another lunch or supper),
half a meal, or a quarter of a meal. Afull meal means exactly the
amounts of carbohydrates and protein that you would ordinarily eat
at that meal. Halfa meal means exactly halfthe protein and halfthe
Even ifyour blood sugar has notyet come back totarget, since you
knowyou have consumed the proper amount ofglucose toeventually
bringit back, youcanconfidentiyinject theamount of insulinor swal
low the dose of the OHA that younormally use to cover that meal.
Forhalfa meal, take halfthe dose; for a quarter of a meal, take one-
quarter the dose.
Don'tfrustrate yourself bywaiting theusual 45 minutes or soafter
injecting regular insulin, or the20 minutes after injectinglispro (oras
part or glulisine), or the60-120 minutes after taking an OHA. Just in
ject and eat. An extra meal now and then won't make you fatter or
cause harm. Since you're eating within the controlled boundaries of
your meal plan and not gorging on sugars or unlimitedamounts of
food, you'restillabidingbythe Laws of Small Numbers.
Ifyou know how much insulin or OHAyou usually take to cover a
certain snack, you might have the snack instead of the meal.
Two ofthemost common effects ofhypoglycemia can make thejobof
helping you difficult andunpleasant. These effects are irritable, nasty
behavior andfailure to recognize your own symptoms. Atmyfirst in
terview with many new patients and their families, instances of vio
lence during hypoglycemic episodes are commonly reported. The
most common scenario I hear goes likethis:"WheneverI seethat he's
low, I hand hima glass of orange juice andtell himto drink it, but he
throws the juice at me. Sometimes he throws the glass too." Such sto
ries comeas no surpriseto me because as a teenager I usedto throw
the orangejuiceat my mother, and when I wasfirst married, I did the
330 Treatment
same tomywife. Whydoes this happen, and howcan we prevent such
First, it's important to try to understand what's going on in the
minds of you and the family member or partner during about with
hypoglycemia. The cognitive difficulties that accompany severe hypo
glycemia can make the slightest frustration or irritation overwhelm
ing. Your low blood sugar may cause you to act bizarrely, as if
intoxicated — andinasense, youare intoxicated. Because yourthink
ing isimpaired, you may betotally unaware that your blood sugar is
low. The similarity to drunkenness is not a coincidence, since the
higher cognitive centers of thebrain, which control rational behavior,
are impairedin both cases.
You probablyhave learned that high blood sugars are tobeavoided,
andat somelevel, youremember this, perhaps even cling to it, despite
your hypoglycemia. If someone tries to cajole youintoeating some
thing sweet, you may decide that it's theother person who's irrational.
This is especially trueif the other person has done the same thingin
the past, when blood sugars were actually normal or even high. In
"self-protection" against thesupposed irrational attemptto get youto
eat something sweet, you instinctively may become violent. Most
commonly, thisoccurs if anattempt is made to put food or drinkin
your mouth. You might view this as an"attack." In less rational mo
ments, you mayeven decide, since youknowthat high bloodsugars
are harmful,that yourspouse or relative istryingto kill you.
The helping relative, usually aspouse or parent, maybe terrified to
seesuch strange behavior. If your loved one hasbeen through many
such encounters, he or she may, for self-protection, keep candies or
other sweets aroundthe housein the hopesthat youwill eatthem and
thus avoidsuch situations. The fear canbe exacerbated if your loved
one has seen you unconscious from hypoglycemia, or ismerely aware
that hypoglycemia can cause dire consequences. On otheroccasions,
whenyour blood sugar wasn't really low, your loved onemayhave er
roneously asked you to eat something sweet. Such erroneous diag
nosesare especially common during family squabbles. The spouse or
parent mayfeel that"his blood sugar islow, and that's whyhe's yelling
at me."Your loved one would rather playit safe and give you some
thingsweet, evenif yourbloodsugar isn'tlow.
There is a solutionto this apparent dilemma. First of all, both par
tiesmust recognize that, as arule, about halfthe time that the relative
Howto Preventand Correct LowBloodSugars 331
suspects hypoglycemia, you do not havea lowblood sugar; the other
half of the time, blood sugar is indeedlow.
No one hasevercontradicted me when I'vemadethis point.
Encouraging a diabetic to eat sweets when hypoglycemia is sus
pected, despite conventional teaching, does as much harm as it does
good.A better approach would be for the lovedone to say, "I'm wor
riedthat your blood sugar may be low. Please checkit andlet me know
the result sothat I'll feel less anxious." As a patient, you should realize
that living with a diabetic can often be as much or more of a strain
than having diabetes. You, the diabetic, owe some consideration to the
needsof your loved ones. Tryto look upon the request to check your
bloodsugar not asanintrusionbut as yourobligation to relieve some
one else's fear. With this obUgation in mind, you shouldautomatically
checkyourblood sugar if asked, just to makethe other personfeel bet
ter. It doesn't matter whether your blood sugar is low or normal. If
your blood sugar is low, you can correct it and find out why. If it's
normal, then you probablywill havediffused the tension of the situa
tion, and now you'll be able to get back to whatever you were doing,
unworriedthat blood sugar isofftarget. When youlook at blood sugar
as something like a clock that you can set — and reset — you take
some of the mystery out of it, and candiminish the emotion involved.
If you're without your meter, take enough glucose tablets to raise
your blood sugar about 60 mg/dl — again to make the other person
feel better. This is the least you can do for someone who may worry
about you every day.
Believe it or not, this simple approach has worked for me and for
many of my patients. As I've saidpreviously, I went through this with
my parents and have gone through it with my wife. Spouses report
that it relieves them of a great burden. Some wives have even cried
when expressingtheir gratitude.
This more serious hypoglycemic state is often characterized by ex
treme tiredness and inability to communicate. You may be sitting and
bangingyour handon atable, walking aroundin adaze, or merely fail-
332 Treatment
ingto respondto questions.It's important that those who liveor work
with you learn that this is a fairly severe stage of hypoglycemia. The
likelihoodthat it's hypoglycemia is so great that valuable time may be
wastedif treatment is delayed whilesomeone fumbles about trying to
measure your blood sugar. It's quite possible that if you're given glu
cose tablets you will not chewthem, and may even spit them out.
The treatment at this stage is glucose gelby mouth.
Glucose prepared asasyrupy gel is soldin the United States under
several brandnames.At least one of theseproductsis not pureglucose
(dextrose) but contains a mixture of long- and short-acting sugars,
and therefore will not exertits full effectasrapidlyaswe'dlike.At pres
ent, I ask my patients to purchase a product called Glutose 15 (Pad
dock Laboratories, Minneapolis, MN 55427; available from Rosedale
Pharmacy). Glutose 15is packaged in a plastic tube (like toothpaste),
with a twist-off cap. Each tube contains 15 grams of glucose. From
Table 20-1 (page 325), we see that this amount will raise the blood
sugar of a 140-pound person by 75 mg/dl (15 x 5). An appropriate
dose for most adults in this condition would be 1V4 tubes. These
would typicallyraise one'sblood sugar by about 110 mg/dl.
Some of the tubes of decorative icing used to write on birthday
cakes contain almost pure glucose (dextrose), so you might save
money by purchasingthose. Look in the baking section of most su
permarkets, but make sure of the contents and weight. To convert
ounces to grams, multiply by 30. Make sure that the major ingredient
is glucose, as some brands aremosdy sucrose, which works too slowly.
We recommend that two tubes of Glutose 15,secured together with
a rubber band, be placed at strategic locations about the house and
place ofwork, aswell asin luggage when you travel with acompanion.
It should not be refrigerated, asit may hardenwhen cold. To adminis
ter, someone should insert the tip of an open tube into the corner of
your mouth, in between your lower gum and your cheek, and slowly
squeeze out a small amount. You will probably swallow this small
amount. After you swallow, a bit more of the gel should be gently
squeezed fromthe tube. Within 5 minutes of ingesting, you should be
able to answer questions.
When you have fully recovered, check and correct your blood sugar
to your target.Sinceyou may havewiped the sticky geloff your mouth
with your hands, you shouldwashthem beforestickingyour finger.
Although glucose gels may not be available in many countries, they
are available on the Internet. Most industrialized nations have phar-
HowtoPrevent andCorrect LowBlood Sugars 333
macies and surgicaldealers that sell flavored glucose drinks to physi
cians for performing oral glucose tolerance tests. These are usually
botded in 10-ounce(296ml) screw-top bottiesthat contain 100grams
of glucose. A dose of 2 fluid ounces (60 ml) will provide about 20
gramsof glucose, enoughto raisethe bloodsugarof a 140-poundper
son by 100mg/dl. Tiny amounts can be administeredwith the help of
a plasticsqueeze bottle. Whoever feeds you the liquid or gel must ex
ercisecaution, as the possibilityexiststhat you could inhale some of it,
causingyou to choke. The use of liquid is potentially much more haz
ardous than the gel in this respect, so administer only a tiny amount
for each swallow.
Hypoglycemia is not the only cause of loss of consciousness. Stroke,
heart attack,a sudden drop in bloodpressure, and evena bump on the
headcanrenderyouunconscious. In fact, veryhighbloodsugar(above
400mg/dl) overseveral days, especiallyin a dehydratedindividual, can
alsocauseloss of consciousness. We willassume, however, that if you
are carefully observing the treatment guidelines of this book, youwill
not allowsuch prolongedbloodsugarelevation to occur.
If you're found unconscious bysomeone who knows howto rapidly
check your blood sugar, a measurement may be made. Treatment
should not be delayed, however, while people are scampering about
trying to find your testingsupplies.
The treatment under these conditions is injection of glucagon, a
hormone that rapidlyraises bloodsugarbycausing the liver and mus
cles to convert stored glycogen to glucose. It is imperative, therefore,
that thosewholive withyouknowhowto give an injection. If you use
insulin,you can give themsomepractice byteaching them howto give
you insulininjections.Glucagon is soldin pharmaciesin manycoun
tries as the GlucagonEmergency Kit. This consists of a small plastic
box containing a syringe filled with an inert waterlike solution and a
litde vial of white powder (glucagon). The kit also contains an illus
trated instruction sheet that your family should read before an emer
gencydevelops. The user injects the water into the vial, withdraws the
needle, shakes the vial to dissolve the powder in the water, and draws
the solution backinto the syringe. The tip of the long needle must be
334 Treatment
submerged in the liquid. For adults, the entire contents of the syringe
should be injected, either intramuscularly or subcutaneously; lesser
amounts should be used for small children. Anyof the sites shown in
Figure 16-1 on page251 can be used, as can the deltoid muscle (page
309) or even the calf muscle. Your potential benefactors should be
warned that if they choosethe buttocks, injection should go into the
upper outer quadrant, so as not to injure the sciaticnerve. An injec
tion maybe given through clothingprovidedit isnot too thick (for ex
ample, through a shirtsleeve or trouser leg, but not through a coat or
Under no circumstances should anything be administered by
mouthwhileyou are unconscious. Since you will not be ableto swal
low, oral glucose could asphyxiate you. If your glucagon cannot be
found, your companions shoulddial 911 (in the United States)for the
emergency medicalservice, or you should be taken to the emergency
room of a hospital.
When an individual has lost consciousness from hypoglycemia, he
may experience convulsions. Signs of this include salivation, tooth-
grinding, and tongue-biting. Although the last can cause permanent
damage in the mouth, no attempt to intervene shouldbe made. Your
heroic savior will not be able to help you if you bite off her fingers. If
possible, you should be turned to lieon your sidewith your head po
sitioned so that your mouth is downward. This is to help drain excess
salivafromyour mouth so you won't breathe it in and choke.
You should begin to showsigns of recovery within 5 minutes of a
glucagon injection. You shouldfully regain consciousness and be able
to talk sensibly within 20 minutes at most. If steadyimprovement is
not apparent during the first 10 minutes, the only recourse is the
emergency squad or hospital. The emergency squadshouldbe asked
to inject 40 cc of a 50 percent dextrose (glucose) solution into a vein.
Individualsweighing under 100pounds (45kilograms) should receive
proportionately smaller amounts (e.g., a 70-pound child would re
ceive 20 cc of the dextrose solution).
Glucagon can cause retching or vomiting in some people. Your
head should therefore be turned to the side so that if you do vomit,
you won't inhale the vomitus. Keep a 4-ounce (120 ml) bottie of
metoclopramide syrup on hand, attached with a rubber band to
the Glucagon Emergency Kit. One gulp of metoclopramide, taken
after you are sitting up and speaking, should almost immediately
Howto Preventand Correct LowBloodSugars 335
stop the feeling of nausea. Do not consume more than one gulp, as
large doses can cause unpleasant side effects (see page 368). In the
United States, metoclopramide is available onlyupon prescription by
a physician.
One doseof glucagon canraise your bloodsugarbyas much as 250
mg/dl, depending upon howmuchglycogen was storedinyour liver at
the timeof the injection and subsequendy converted to glucose. After
you've fully recovered yoursenses, youshould checkyourbloodsugar.
If at least 5 hours have elapsed since your last dose of a rapid-acting
insulin, take enough intramuscular (or subcutaneous) lispro (or as
part or glulisine) insulin tobringyourbloodsugar backdownto your
target. This is important, because if your blood sugar is kept normal
for about 24hours,your liver will rebuild its supply of glycogen. This
glycogen reserve is of great value for protection from possible subse
quent hypoglycemic events.
Bythe way, if we tried to give glucagon to someone twice in the
same day, thesecond shotmight not raise blood sugar. This ispossible
because liver glycogen reserves may have been totally depleted in re
sponseto the firstinjection. Thus,monitoringand correctionof blood
sugar every5 hours for 1full dayis mandatoryafter the use of gluca
gon. Additional blood sugar measurements should be taken every
2l/z hoursto make surethat you're not again hypoglycemic, but do not
correct for high blood sugars every 2l/i hours; wait the full 5 hours
sincethe last shot of rapid-actinginsulin (see page305).
Although reading about possible loss of consciousness may be
frightening, remember that thisisanextremely rareevent, andusually
resultswhena type 1diabeticmakes a major mistake, suchasthose in
cluded in the list on pages 319-320.1 know of no case where a type2
diabetic experienced severe hypoglycemia whenusing anymedication
that we recommend.
Thesigns of hypoglycemia duringsleep include cold, clammyskin, es
pecially on the neck, erratic breathing, and restlessness. It certainly
helpsto havealight sleeper sharingyourbed.Parentsshouldcheckdi
abeticchildrenat night and should feel their neck.
336 Treatment
Review your Glucograf data sheet after all hypoglycemic episodes,
even mild ones. It's important that you reconstruct the events leading
up to any episode of lowblood sugar, even if it caused no notable
symptoms. This is one of the reasons whywe recommend (page 76)
that most insulin-takingdiabetics keep faithful records of data perti
nent to their blood sugarlevels and whywewent into so much detail
in Chapter 5 teaching youhowto recordthe information. Since severe
hypoglycemia can leadto amnesia for events of the prior hour or so,
habitual recording of relevant data can be most valuablefor this sce
nario. It is certainlyhelpful to record times of insulin shots, glucose
tablets, meals, and exercise, as well as to note if you overate or under-
ate, and so on. Recording blood sugar data alone may not help you to
figure out what caused a problem. If you experience a severe hypo
glycemic episode or several mildepisodes and cannot figure out how
to prevent recurrences, read or showyour Glucograf data sheet to
your physician. Your doctor maybe able to think of reasons that did
not occur to you.
Keeping Hypoglycemia Supplies
Glucose tablets, glucose gel (Glutose 15), and glucagon can each po
tentiallysave your life. Theywon't help if they'renot around or are al
lowed to deteriorate. Here are some basic rules:
• Place supplies in convenient locations around your house and
• Showothers whereyour supplies are kept.
• Keep glucose tablets in your car,pocket, or purse.
• When traveling, keep a full set of supplies in your hand luggage
and also in your checked luggage — just in casea piece of lug
gageis lost or stolen.
• It may be wise to replace glucagon on or beforethe expiration
date on the vial. In an emergency, however, it isn't necessary for
your savior to worry about the expiration date. In the United
States, glucagon isusuallysoldwithveryshort dating. Manypeo
ple are soldcostiy emergency kits markedwith expirationdates
HowtoPrevent andCorrect LowBlood Sugars 337
To make sureyouarenot caught unprepared bylow bloodsug
ars, you should always keep the following supplies on hand at
both your home and yourworkplace:
If You Take OHAs or Even ISAs
• 1-3 bottles(100 tablets each) Dextrotabs or other glu
cose tablets; always carryglucose tablets withyou
If You Take Insulin and Do Not Live Alone, You Also
• One package of three tubes Glutose 15
• Glucagon EmergencyKit
• 4-ounce bottle metoclopramidesyrup
only a few months later. Don't worry. Glucagon is sold as a
freeze-dried powder that will probablyremain effective for five
years after the "expiration"date, unless of course it has been ex
posed to moisture or extreme heat (as in a closed car in the sum
mertime). It retains its longevity especiallyif it is refrigerated.
Once diluted, however, it is good for only 24hours.
• Always replace supplies when some havebeen used. Never allow
your stock to become depleted. Keep plenty of extra glucose
tablets and blood sugar test strips on hand.
Emergency Identification Tags
If youuseinsulinor OHAs, youshouldwear an identification tagthat
displays a recognizable medical emblem, such as a red serpent encir
clinga red staff. The tag,whichmaybe worn as a braceletor necklace,
shouldbe engraved witha message that relates to the treatment of hy
poglycemia. Myownbracelet is engraved withthe following message:
SCIOUS — to hospital. Since bracelets are more likely to be spotted
by emergency personnel, I prefer them to the necklaces.
Most pharmacies and jewelers sell medical ID tags. Prices begin at
338 Treatment
$5for stainlesssteeland gointo hundreds of dollars for solid gold. The
MedicAlertFoundation, Turlock, CA95381, will keep a record of your
medical historyand will sendyoua stainless steelIDbracelet or neck
lace,with their emblem,for $45.Sterling silver or gold-platedIDs cost
slightly more. Beautiful 14-carat goldIDsareavailable at considerably
higher cost.Theywillalsoengrave the tagdiabetic for the samecost.
Alltagsare stampedwith your special IDnumber and with their "call
collect" 24-hour telephonenumber. Byphoning this number, a hospi
tal can secureyour name and address, contact information for your
next of kin and physician, a list of allyour medicalconditions, and the
doses of medications that you take. You can obtain an application
form bywritingto the above address or byphoning (800) id-alert.
Diabetics who do not take medications that can cause hypo
glycemia would alsobe wiseto wear a MedicAlert bracelet, if only to
discourage the automatic use of intravenous glucose infusions — a
common practiceof emergency personnel on victims of motor vehi
cle accidents, heart attacks, and so on.
Emergency Alarm Service
If you livealone, you may want to consider using an emergency alarm
system.These can automaticallyphone a friend, relative, or emergency
squad when you push a button on a necklace. The system can also be
activatedif you do not "checkin"at predetermined time intervals. The
least expensive system that I have encountered is supplied by the
MedicAlertFoundation. Their "failureto checkin"alert unfortunately
can only be activatedat 24-hour intervals,so you can be unconscious
for 24 hours before someone is notified.
The Continuous Glucose Monitor
Most of us have jobsthat bringus into contactwithother peopleduring
thedayandfamilywithwhomwehave contact afterwork. These contacts
offer considerable protection from severe hypoglycemia, as colleagues
and relatives willintervene ifyoustart walkingintowalls or talkingsilly. A
sleeping partner can frequendy pick up on the labored breathing and
cold, clammyskin or damp nightclothes that accompany hypoglycemia
and then awaken youand askyouto checkyour blood sugar.
If you liveor sleepalone, or if your sleepingpartner is an extremely
HowtoPrevent andCorrect LowBlood Sugars 339
deep sleeper, however, youdon't have thisprotection at night.* A new
backup is now available.
Several companies are now marketing continuous blood glucose
monitors.1 A continuous blood glucose monitor works via a tiny sen
sorimplantedbeneaththe skin, usingatechnique similar to that used
for insulin pump tubing.The sensor constantly measures glucose con
centrationin the tissue fluid present at its subcutaneous location. A
combinedpower supplyandradio transmitter attaches to yourskinor
clothing. The transmitter sends up to several hundredglucose read
ings dailyto a small portable receiver that you can keep in a pocket.
The number displayed is approximately equal to the blood sugar
about 20 minutes prior to the reading. So if you had taken a reading
with yourconventional method 20minutesago, thiswouldbe roughly
the sameasthe reading fromthe sensor right now.
Also displayed is an up or down arrow to indicate whether blood
sugar is increasing or decreasing. What's most valuable is an audible
alarm that can be set to sound at any selected blood sugar value and
alsoto signal rapid drops in blood sugar.
There aresome potential problems associated with these devices, so
they're not for everyone, and certainly not for me.
• The sensorremains under the skin for 3 days. During this time,
there is always a possibility for inflammation or infection (prob
ably a low risk).
• Fibrosis, or scartissue, can potentiallybuild up at the sensor site
over time (although how long this would take is as yet unclear)
and eventually make measurements less accurate.
• Measurements are inherently less accuratethan ordinary blood
sugar monitoring, so the devices need to be calibrated against
finger-stick blood sugars about twice daily. For now, at least, I
would recommend that any blood sugar correction be made
based on finger-stick measurements.
• Both the equipment and the disposable sensors with associated
supplies arequite costly and may not be coveredby many insur
ance policies.
*Manyinsulin users will awaken automatically when blood sugar gets too low
during sleep, so these people havebuilt-in protection.
*Search the Webfor "continuous glucose sensor"to comparedifferent models.
340 Treatment
• Advertising for these products may falsely imply (but does not
stateoutright) that a sensor-insulin pump combinationwill au
tomatically monitor and inject insulin to keep blood sugars on
target around the clock.
• The sensor is typically implanted in the abdomen, and has a
• The sensor will typically work for only 3 days, because the en
zymeusedis then depleted.
• The setup requires training to use.
While the sensors have at best limited usefulness at the moment, it
isentirelyfeasible—just fromanengineering perspective—that they
willvastlyimproveovertime.That said,I'm not holdingmybreath —
this technology has beenaround for decades, and manufacturershave
made plenty of moneyby employing it shoddily. One can still hope,
however, that some brilliant entrepreneur will develop a highly accu
rate and timelysensor that can provideconstant, accurateblood sugar
But we'renot there yet.Soto sumit up, if I werelivingalone, I'd use
the sensor to protect from nighttime hypoglycemic episodes and for
get about using an insulin pump.
Some diabetics have absent or diminished ability to experience the
warningsigns of hypoglycemia. Thisoccurs under five circumstances
that have been documented in the scientific literature:
• Severe autonomic neuropathy (injury, by chronicallyhigh blood
sugars, to the nervesthat control involuntary bodily functions).
• Adrenal medullary fibrosis (destruction, by chronically high
blood sugars, of the cells in the adrenal glandsthat produce epi
nephrine). This is especially common in long-standing poorly
controlled diabetes.
• Bloodsugarsthat are chronically too low.
• The use of beta-blocking medication for treatment of hyperten
sion or cardiac chest pain.
• The use of large(nonphysiologic) dosesof insulin, as is common
for individuals on high-carbohydrate diets.
HowtoPreventand Correct LowBloodSugars 341
All ofthese situations result inlowered production of, or sensitivity
to, epinephrine, the hormone that produces tremor, pallor, rapid
pulse, and other signs that we identify withhypoglycemia. It is ironic
that epinephrine production or sensitivity is most commonly dimin
ished in those whose blood sugars have been chronically either very
high or very low.
Injuryto the autonomic nervous system by elevated blood sugar
hasbeendiscussed on pages 61-62. Individuals whose heart ratevari
ation on the R-R interval study is severely diminished may be espe
cially susceptible to this problem.
People who have frequent episodes of hypoglycemia or chronically
lowblood sugar tendtoadapt tothis condition. They appear tobeless
sensitive totheeffects ofepinephrine, which, when repeatedly released
in large amounts, down-regulates its own receptors. This condition
cannot bepredicted byR-R studies. It is, however, readily detectable if
you measure your own blood sugar frequentiy. If causedby chroni
cally low blood sugar, this condition can be reversed bytaking mea
sures to ensure that blood sugaris maintainedat normal levels.
Hypoglycemia unawareness candeprive one of potentially lifesav-
ing warning signals. To compensate for this disability, blood sugar
should bechecked more frequently. For some rare insulin users, it may
be necessary, for example, to measure blood sugar every hour for 5
hours after meals, instead of onlyonceor twice after each meal. For
tunately, we have the tools to circumvent this problem; we need only
to usethem diligendy.
I frequently encounter patients who do not take glucose tablets for
lowblood sugar measurements because they "feel fine." These arejust
the people who are most likely to lose consciousness or find them
selves in an automobile accident.
Whether or not you have hypoglycemia, it is essential that you
check your blood sugar before driving a car and —after finding a
place where you can safely stop your vehicle — every hour while
Syncope, or fainting, is fairly common aspeople get older. It is espe
ciallycommon among diabetics. Even more common isnear-syncope.
342 Treatment
This is merelythe feeling that you will pass out unless you he down
right away. Simultaneously, your surroundings maylook gray or your
vision may fade. Thereare manycauses of syncope andnear-syncope.
These include cardiac and neurological problems,certainmedications,
and dehydration. Thesecauses are not nearly ascommon in diabetics
asare suddendrops of bloodpressure caused by autonomicneuropa
thy or by inappropriate useof antihypertensive medications — espe
cially diuretics ("water pills") and alpha-1 adrenergic antagonists,
such as prazosinand terazosin.
When most of us stand from aseated,supine, or squatting position,
the brain sends a message to the blood vessels in our legs to constrict
reflexively andinstantiy. This prevents blood frompoolingin the legs,
which would deprive the brain of blood and oxygen. If you've had
highblood sugars for manyyears, the nerves that signal the vessels in
the legs may conduct the message poorly (a sign of autonomic neu
ropathy). A drop in bloodpressure upon standing, called postural, or
orthostatic, hypotension, occurs whenthis pooling in the legs occurs.
For some,the heartmaybringbloodpressure backupby increasing its
rateand amount ofcontraction.Unfortunately,this does not occur for
many diabetics with autonomicneuropathy.
Alternatively, if you eat abig meal, blood may concentrate in your
digestive system, also depriving the brain. The normal mechanisms
that protect the brain from this shunting of bloodmaybe deficient if
you have autonomic neuropathy. It is in part to gauge potential for
these reactions that I measure supine and standing blood pressures,
and performR-Rinterval studies on all my diabetic patients. A recent
study of medical (mosdy nondiabetic) outpatients in the United
States suggests that 20percent of individuals overthe age of 65and30
percent of those over age 70 have documentable postural hypoten
sion. Fordiabetics the incidence is probably much greater.
A common scenario for syncope or near-syncope involves the dia
beticwho gets up in the middleof the night to urinate andkeels over
on the wayto the bathroom. A simple wayto avoid this is to sit at the
edge of the bedwith feet dangling for a few minutesbefore standing.
Another syncopescenario involves the personwho goes to the toi
let and passes out while tryingto produce abowelmovement or uri
nate. Again, the reflexes that prevent shunting of bloodaway from the
brain areblunted by autonomic neuropathy.
If syncope iscausedbytransient lowcerebral bloodpressure as are
sult of autonomic neuropathy, one shouldlaythe victim out flat and
Howto Prevent and Correct LowBloodSugars 343
elevate his feet high abovehis head. He should return to consciousness
almost immediately.
The symptoms of syncope are similarto those of moderate to se
vere hypoglycemia. In both cases, the brainisbeing deprived of a ba
sic nutrient — oxygen in the case of syncope, glucose in the case of
hypoglycemia. Furthermore, postural hypotension canalso occurasa
result of hypoglycemia. Some symptoms of near-syncope include
faintness, visual changes, and disorientation.
Whatever the cause of fainting or near-syncope, blood sugar must
be checked to rule out hypoglycemia. If blood sugar is normal, no
amount of glucose will cure theproblem. People with recurrent pos
turalhypotension will usually find reliefbywearing surgical stockings
of 30-40 mm compression. If these are inadequate, waist-high surgi
cal pantyhose should be used.
If you've heard horror stories about thefrequency andseverity of se
vere hypoglycemia in type 1diabetes, thepeople you've been hearing
about are probably taking industrial doses of insulin to cover large
amounts of dietarycarbohydrate. On our regimen, this hazardis vir
tually nil.Someone would have to make a major mistake, suchas tak
ing an insulin dose twice, or not waiting the full 5 hours before
correcting an elevated bloodsugar, forlife-threatening episodes to oc
cur. Many type 1diabetics seek me out because of their frequent hy
poglycemic episodes andnot necessarily because of their high blood
sugars. Our regimen takes care of both.
Please don't neglect toask others toread thischapter. When youare
mostin needof help fortreating hypoglycemia, youmay beincapable
of rendering it yourself. So show this chapter to your close relatives,
friends, and coworkers and ask them to read it. It should increase their
own confidence incoping with such situations, andthepotential pay
off to you maybe considerable.
How to Cope with Dehydration,
Dehydrating Illness, and Infection
When you experience vomiting, nausea, fever, diarrhea, or
anyformof infection, youshouldimmediatelycontactyour
physician. I can't really emphasize enoughthe importance
of getting treatment and getting it fast. To drive home this point, I'll
sharethe following experience.
Some years ago, I got a call from a woman at about four o'clock
on a Sunday afternoon. She wasn't my patient, but her diabetologist
was out of town for the weekend with no backup for emergencies. He
hadnevertaught herwhat I teach my patients — the contents of this
She found my Diabetes Center in the white pages of the phone
book. She was alone with her toddler son and had been vomiting con
tinuously since 9:00 a.m. She asked me what she coulddo. I told her
that she must be so dehydrated that her only choice was to get to a
hospital emergency room as fast as possible for intravenous fluid re
placement. While she dropped off hersonwithhermother, I called the
hospital andtoldthemto expect her. I got acall 5hourslater from an
attending physician. Hehadadmitted her to the hospital because the
emergency room couldn't help her. Why not? Her kidneys had failed
from dehydration. Fortunately, the hospital had a dialysis center, so
they put heron dialysis and gave herintravenous fluids. Had dialysis
not been available, she would likely have died. As it turned out, she
spent five days in the hospital.
Clearly, adehydrating illness is not something to takelightly, not a
reason to assumeyourdoctorisgoing to think you'reahypochondriac
if you call every time you have one of the problems discussed in this
How toCope withDehydration, Dehydrating Illness, andInfection 345
chapter. This issomething that could kill you, andyou need prompt
Why isit, then, that diabetics have a more serious time with dehy
drating illness thannondiabetics? Clearly it has something to dowith
blood sugars.
Ifyou are vomiting or have diarrhea, you've either been poisoned (un
likely) or have an infectious illness. If you have an infection, whether
it's in your mouth, on your finger, or in your gastrointestinal tract,
your blood sugar is most likely going to go up. So you're starting off
with elevated blood sugars just byvirtue ofthe infection. Ifyou vomit
or have diarrhea, you are losing fluid from a region inthe body that
normally contains fluid. That lost fluid is going to be replaced from
the largest source of fluid in the body, the bloodstream. It's not that
you're going to bleed into your stomach —your GI tract is full of
blood vessels that are there in part for the exchange of fluids. That's
how fluid is absorbed.
Your bodynaturallytries tomaintain abalance, sowhen fluid disap
pears from one place, your bodytries toreplace itusingwater from your
bloodstream. But as your blood loses water, glucose isleft behind, and
youend upwitha higherbloodsugar concentration. In addition, blood
vessels area giantweb throughout thebody, but unlike aweb, theves
sels narrow as they travel out from the center, narrowing from inside
thebody to outside, from inside anorgan to itssurface, andsoon. At
any given time, much oftheblood is inthese narrow, peripheral vessels.
If your bloodstream has lost significant amounts of fluid, as you
would inadehydrating illness, the periphery is notgoing tobeas well
supplied asit would normally be. It's like having a whole new insulin
resistance simplybecause insulin andglucose aren't adequately reach
ingthenarrower vessels. Since less glucose will bedelivered to thecells
adjoining these vessels, yourblood sugar concentration will continue
toclimb. Furthermore, the higher your blood sugars go, the more in
sulin resistance you will experience. The more insulin-resistant you
are, thehigher your blood sugars are going tobe. Avicious circle.
To make the circle even more vicious, when you have high blood
sugars, you urinate —andofcourse what happens then isthatyou get
346 Treatment
even more dehydrated and more insulin-resistant and your blood
sugar goes even higher. Nowyour peripheral cells have a choice—
either die from lack of glucose and insulin or metabolize fat. They'll
choose the latter. But ketones are createdby fat metabolism, causing
youto urinateeven moreto ridyourself of the ketones, taking youto
a whole new level of dehydration.
This sequence of events can happen in a matter of hours, as it did
with the womanjust described. Sothe nameof the gameisprevention.
Howdoyouprevent illness from causing dehydration? Let's sayyou
wake up in the middle of the night or in the morning and vomit or
have a bout of diarrhea. What do you do? Call your physicianand let
him or her know—even if it's two o'clockin the morning, call your
doctor. Even if it turns out to bejust somethingyouateand it's a tran
sient episode, callyour doctor or his/her answering service.
We allgetsickfromtimeto time, but ifyou're on our diet and treat
ment plan, and if you'rereasonably healthy, youshouldn't get sickany
more frequently than the average person—and probably less fre
quently thanthe average diabetic. For diabetics, however, such illness
can pose specialproblems.
As you know, sickness or infection can cause your bloodsugar to
increase, and injected insulin—even if you don't normally take in
sulin—can help preserve beta cell functionduring illness (as well as
help keep your blood sugar under control and thereby reduce de
hydration). One of the most pressing concerns for diabetics during
illness is dehydration, which, as illustrated above, can lead to life-
threatening consequences if not handled effectively and rapidly.
Commoncauses of dehydration include not onlymultiple episodes of
diarrhea or vomiting, and fever and resulting perspiration; theyalso
includefailure to drink adequate fluids, especially duringhot weather
or prolonged exercise, andvery high blood sugars. You probablyknow
that one of the hallmarksymptoms of veryhigh blood sugars is the
combination of extreme thirst and frequent urination. From what
you've already read in this chapter, you should understand the equa
tion. Still, I think it's noteworthy enough to lay it out again for em
HowtoCope withDehydration, DehydratingIllness, andInfection 347
1. Dehydrationcauses transitoryinsulin resistance.*
2. During periods of dehydration, bloodsugar will tend to rise.
3. High blood sugar, as you know, itself leads to insulin resistance
and further blood sugar increase.
4. Blood sugar elevation from dehydration in addition to blood
sugar elevationcausedbythe viral or bacterial infection that led
to your vomiting, fever, or diarrhea causes further insulin resis
tance and blood sugar elevation.
5. High bloodsugarcauses further dehydration asyour kidneys at
tempt to unloadglucose andketones byproducing large amounts
of urine.
6. Increased dehydration causes higher blood sugars, which in
turn cause further dehydration. All of whichbrings us back to
number 1.
The good news is, however, that simple interventions can halt this
spiraUng of bloodsugars and fluid loss. It'sthepurposeof thischapter
to give you the knowledge to prevent the sort of grave consequences
experienced bythe lady whocalled me on that Sunday afternoon —
or worse, death.
There are two acute conditions that can develop from the combina
tion of high blood sugarsand dehydration. The first is called diabetic
ketoacidosis, or DKA. It occurs in people who make virtually no in
sulinon their own(eithertype1diabetics or type2diabetics whohave
lost nearly all beta cell activity). Very lowserum insulin levels, com
binedwiththe insulin resistance caused byhighbloodsugars and de
hydration, result in the virtual absence of insulin-mediated glucose
transport to thetissues of thebody. Intheabsence of adequate insulin,
thebodymetabolizes storedfats to produce the energy that tissues re
quire to remain alive. Aby-product of fat metabolismis the produc
tion of substances called ketones and ketoacids. One of the ketones,
*It is absolutely important when experiencing dehydrating illness not to do any
thingthat wouldhastendehydration —and that indudes the useof certainmed
ications, such as ACE inhibitors and diuretics. Never discontinue a medication
without discussing it with your doctor, soask your physician assoon asyou experi
ence such anillness aboutceasing use ofthese andsimilar medications. Ifyou're un
sure, mostpharmacists cantell youifa particular drugcanfacilitate dehydration.
348 Treatment
acetone, is familiar as the major component of nail polish remover.
Ketones maybe detected in the urinebyusing a dipstick suchas Ke-
tostix (see Chapter 3, "Your Diabetic Tool Kit"). Ketones may alsobe
detected on the breath as the aroma of an organic solvent, which is
why unconscious diabetics areoften mistaken for passed-out drunks.
Ketones and ketoacids are toxic in large amounts. More important,
your kidneys will tryto eliminate themwitheven moreurine,thereby
causing furtherdehydration. Some of thehallmarks of severe ketoaci
dosis are large amounts of ketones in the urine, extreme thirst, dry
mouth, nausea, frequent urination, deeplaboredbreathing, and high
blood sugar (usually over350mg/dl).
The other acutecomplication of highbloodsugar and dehydration,
hyperosmolar coma, isa potentially more severe condition, and occurs
in people whose beta cells still make some insulin. ("Hyperosmolar"
refers to high concentrations of glucose, sodium, and chloride in the
blood due to inadequate water to dilute them.) Diabetics who develop
this condition usually have some residual beta cell activity, making
enoughinsulin to suppress the metabolism of fats, but not enoughto
preventveryhighbloodsugars. As a result, ketones maynot appearin
the urine or on the breath. Because this condition most commonly oc
curs in elderly people, who do not become very thirsty when dehy
drated, thedegree ofdehydration isusually greater thaninketoacidosis.
Early symptoms of a hyperosmolar state include somnolence and con
fusion. Extremely highbloodsugars (asgreat as 1500 mg/dl) have been
reported in cases of hyperosmolar coma. Fluid deficit maybecome so
severe that the brain becomes dehydrated. Loss of consciousness and
death can occur in both the hyperosmolar state and in severe DKA.
The treatment for DKA and hyperosmolar coma includes fluid re
placement andinsulin. Fluid replacement alone canhave a great effect
uponbloodsugar because it bothdilutes theglucose level in theblood
and permits the kidneys to eliminate excess glucose. Fluidalso helps
the kidneys eliminate ketones in DKA. Our interest here, though, is
not in treating theseconditions —this must bedonebya physician or
in a hospital — but in preventing them.
Vomiting, nausea, and diarrhea aremost commonly caused bybacte
rial or viral infections sometimes associated with flulike illness. An es-
How toCope with Dehydration, DehydratingIllness, andInfection 349
sential part of treatment istostop eating. Since you can certainly sur
vive a few days without eating, this should pose no problem. But if
you're noteating, it makes sense toask what dose ofinsulin orISAyou
should take.
Adjusting Your Diabetes Medication
If you're on one of the medication regimens described in this book,
theanswer issimple: you take theamount andtype ofmedication that
you'd normally take to cover the basal, or fasting, state and skip any
doses that are intended tocover meals. If, for example, you ordinarily
take detemir or glargine as basal insulin upon arising andat bedtime,
and regular or lispro (or aspart or glulisine) insulin before meals,
you'd continue the basal insulin and skip the preprandial regular or
lispro for those meals you won't beeating. Similarly, ifyou take anISA
on arising and/orat bedtime for thefasting state, andagain to cover
meals, you skip thedoses for those meals that you do not planto eat.
Inbothof theabove cases, it's essential that the medications usedfor
the fasting state continue at their full doses. This is in direct contradic
tiontotraditional "sick day" treatment, butit's amajor reason why pa
tients who carefully follow our regimens should not develop DKA or
hyperosmolar comawhentheyareill.
Ofcourse, if you're vomiting, you won't beable to keep down oral
medication and thisposes yetanother problem.
Remember, because infection and dehydration may each cause
blood sugar to increase, you may need additional coverage for any
blood sugar elevation. Such additional coverage should usually take
the form of lispro insulin. This is one of the reasons that we advocate
the training of all diabetics in the techniques of insulin injection —
even those who, when not sick, can becontrolled byjustdietandISAs.
Using insulin when you're sick may beespecially important for you,
becauseit helps to relieve the added burden on beta cells that leads to
burnout. This isbut oneofthereasons it's mandatorythatyou contact
yourphysician immediatelywhen you feel ill. Heor sheshould beable
totell you howmuch coverage with lispro will benecessary, and when
to take it. The protocol for such coverage isdiscussed on pages 301—
304, butbecause ofits importance, it bears repeating again briefly:
1. Measure bloodsugars on arising and every 5 hours thereafter.
2. Inject enough lispro at these times to bring your blood sugars
down toyourtarget value. Intramuscular shots arepreferred (see
350 Treatment
pages 307-310) because of their rapideffect, but subcutaneous
injection isalso acceptable. It isprudent to continue bloodsugar
measurementsand insulincoverage, evenduring the night, for as
long as blood sugarscontinueto rise.
If you're soillthat youcannotcheck yourownbloodsugars and in
ject your own insulin, someone else must do this for you, or you
shouldbe hospitalized. Thepotential consequences aresoseriousthat
you haveno other options.
Medications to Be Discontinued
Certain medications that can accelerate dehydration or temporarily
impairkidney function should bediscontinued duringa dehydrating
illness. These include diuretics, ACE inhibitors, and certain arthritis
medications such as NSAIDs (ibuprofen, Motrin, Advil) and COX-2
inhibitors.NSAIDs may, however, be usedasa last resort to treat fever
if other medications are ineffective. Discuss this with your physician
beforediscontinuing medication he has prescribed. If you can't reach
him, then discontinue.
Controlling the Vomiting
The mainstay of treatment is fluid replacement, but if you've been
vomiting, you'll probablybe unable to holdanything down,including
fluids. If symptoms disappear aftervomiting once and you can keep
things down, then there's likely no needfor treatmentto prevent fur
ther vomiting {butstill notify your physician). Ordinary vomiting can
usuallybesuppressedwithTigan (trimethobenzamide hydrochloride)
suppositories, administered rectally every 3-5 hours if vomiting per
sists. Tigan should not betaken bymouth,asit will probablybevom
ited up before it can work. Suppositories should be stored in your
refrigerator, as they tend to melt in hot weather. The suppository
should be inserted well into the rectum, blunt endfirst, so it won't
come out.
Tigan works for most people, but in about a third of cases, it
doesn't, which is all the more reason to contact your physician when
youexperience apotentiallydehydrating illness (nausea, vomiting, di
arrhea, fever). If vomitingor nauseacontinuesfor more than 4 hours,
or if it cannot be haltedbyTigan within 1 hour, he or she maywant
you to try a second or even a third dose or may prescribe a visit to a
hospital emergency roomto receive intravenous fluid (saline) and to
How toCope withDehydration, DehydratingIllness, andInfection 351
have the cause established. Some surgical emergencies such as intesti
nal obstruction can lead tovomiting, as can poisoning, gastroparesis
(Chapter 22), DKA, and so on. Vomiting is a serious problem for
people withdiabetes, andshould not betreated casually.
Large doses of Tigan cancause bizarre neurological side effects, es
pecially in children and in slim elderly people. When vomiting has
ceased, it should probably notbeadministered more often than every
3hours, or indoses greater than that prescribed by your physician. I
usually recommend 100 mg suppositories for children and slim
people olderthan 65years, and200 mgsuppositories for all others. If
Tigan doesn't work fully within 1hour, take more andcall your physi
cian again.
Fluid Replacement
Once vomiting has been controlled, you should immediately begin to
drinkfluids. Two questions naturally arise at this point: What fluid?
And how much? There are three factors that must be considered in
preparing the fluid to be used.
First, it mustbepalatable. Second, it should contain nocarbohydrate
(therefore noGatoradeor other sportsdrinks),but artificial sweeteners
areokay. Thisguideline alsocontradicts conventional treatment, which
usuallycalls for sweetened beverages to offset the excessive amountsof
insulin that many diabetics use. Finally, the fluids should replace the
electrolytes — sodium, potassium, and chloride— that are lost from
thebodywhen we lose fluids. Beverages commonlyusedbymypatients
include diet soda, diluted iced tea, seltzer, water, and carbohydrate-
free bouillon or clear soup. To these fluids, weaddelectrolytes.
To each quart of liquid, add:
Exactly but no more than 1 level teaspoon table salt (Vi tea
spoon if it tastestoo salty) (provides sodiumand chloride)
Exactly but no more than V4 teaspoon salt substitute (see list,
page 67) (provides potassium and chloride)
Ifthevomiting ceased after oneepisode without theneed forTigan,
it isn't necessary to addthesalts to thefluid youconsume.
In anticipation of these rare "sick days," you should always have
on hand several 2-quart bottles of diet soda or seltzer, or two empty
2-quart plastic iced tea pitchers. The pitchers can be used to store
whatever rehydration concoction youmayprefer instead of diet soda.
352 Treatment
When theneedarises, onepitcher of fluid can bekeptby yourbedside,
whilethe second is kept coolin the refrigerator.
Thevolume of fluid youwill require each day, whennot eating, de
pends upon your size, since large people utilize more fluid than small
people. If your blood sugars are elevated orif your urine ondipstick is
positive for more than moderate amounts of ketones, you will need
much more fluid than otherwise. The ongoing fluid requirement for
most adults without these problems comes to about 2-3 quarts daily
while fasting.* In addition, within the first 24 hours you should re
place the estimated fluid loss caused by vomiting, fever, or diarrhea.
This maycome to another few quarts, soclearly youwill have to do a
lot of drinking. Your physician should be consulted for instructions
regarding yourfluid intake while ill. If for anyreason youcannot con
sume orkeep down theamount of liquid that she orherecommends,
you may have tobehospitalized toreceive intravenous fluids.
If youdohave tobehospitalized for IV fluid replacement, youmay
run into the difficultyof inexperienced or ignoranthospitalpersonnel
wanting to give you one or another standard solution that contains
somesort of sugar — glucose, lactose, lactated Ringer's solution, fruc
tose, and so on. Do not allowthem to do so, and do not assume they
know more than you do aboutyour situation. Insist upon asaline so
lution,1 and if they balk, insist upon speaking with the hospital ad
ministrator and threaten malpractice and wrongful death lawsuits, if
necessary, to persuade them of what you need. Although not usually
effective outside the United States, such threats are usually effective
herebecause malpractice insurance companies will likelyinsist upon
the discharge of the person who precipitates alawsuit.
Here again we are faced with three basic problems: blood sugar con
trol, control of the diarrhea to prevent further water and electrolyte
loss, and fluid and electrolyte replacement.
The guidelines for blood sugar control are the same as if youhave
*Figure on 0.02 quarts per pound of body weight (0.044 liters per kilogram).
Thus,about 3quarts(or 3liters) for a person weighing 150 pounds(68kilos),
beware of D-5 or D-10 salinesolutions. These contain dextrose (glucose) and
will certainly raise yourblood sugar andthereby cause furtherdehydration. For
uncontrollable diarrhea, halfnormal saline should be used; otherwise, normal
How toCope withDehydration, DehydratingIllness, andInfection 353
been vomiting (see above). Fluid and electrolyte replacement should
be the same as for vomiting, except that 1level teaspoon of sodium
bicarbonate (baking soda) should be added to each quart of the
electrolyte-replacement mixture. The primary treatment for diarrhea,
as for vomiting, is to stop eating. Medications to relieve diarrhea, if
any, should be specified by your physician. Some forms of diarrhea
caused by bacteria, such as "traveler's diarrhea," maywarrant the use
of Pepto-Bismol (bismuth subsalicylate) and antibiotics such as cipro
In my experience, there is oneantidiarrheal agent that has always
worked, Lomotil (diphenoxylate Hcl withatropine sulfate). This is a
prescription drug thatyou should have your doctor prescribe inliquid
form in adropper bottle (in advance of any illness). The generic ver
sions are muchless expensive and just as effective. You should always
have several bottles on hand. You will find dosing instructions onthe
package insert. If diarrhea continues, double thedose every hour un
tilit ceases and continue the final dose every 3hours until your physi
cianadvises you to discontinue. (Oncethe diarrhea ceases, it would be
moreconvenient andcheaper for an adult to switch from theliquid to
the tablet form of the drug. One 2.5 mg tablet is equivalent to 1tea
spoon or full dropper of theliquid.) Overdosing will not onlydryout
your gut, which we are seeking, but can dry out your larynx, mouth,
nose, andeyes. Lomotil can also make youdrowsy, but itseffect on di
arrhea is miraculous, in my opinion.
No doubt you've heard the advice, "Drinkplentyof fluids," for a fever.
This isbecause fever causes considerable fluid loss through theskin as
perspiration. Your loss of fluid can be difficult to estimate, so your
physician maywant to assume that you'd require 1-2 morequarts of
fluid daily than you'd normally need. Ordinarily, amildfever helps to
destroy the infectious agent (virus or bacteria) that caused the fever.
The tendencyto sleep out fever mayalso be beneficial. For adiabetic,
however, thesomnolence that you experience withfever maydiscour
age you from checking yourbloodsugar, covering with insulin, drink
ingadequate fluid, and calling your physician every few hours. If you
don't have someone awaken youevery 20minutes, youshould useas
pirin, acetaminophen (Tylenol), or ibuprofen (Advil or Motrin), in
354 Treatment
accordance withyourdoctor's instructions, tohelpfight thefever. Be
ware, however, that aspirin can cause false positive readings on tests
for urinary ketones, sodon'teven test for ketones if you are using as
pirin. Never use aspirin or ibuprofen (or any ofthe nonsteroidal anti
inflammatory drugs, NSAIDs) for fever in children because of the risk
of Reye's syndrome. Excessive doses of aspirin or NSAIDs (naproxen,
ibuprofen, and many others) can cause severe hypoglycemia. If at all
possible, try not to use NSAIDs, as the combination of these drugs
with dehydration can cause kidney failure. Acetaminophen can be
highly toxic if used indoses greater than those indicated onthe pack
If you have fever, the guidelines for blood sugar control and re
placement of fluid are almost the same as indicated previously for
vomiting. There is one difference, however. Since there is very little
electrolyte loss in perspiration, it's not necessary to add salts to the
fluid you consume if you're not vomiting or experiencing diarrhea.
Certainly there isnoreason nottoeat ifyou feel hungry—but if you
wanttoeat,cover yourmeals withyourusual dose of insulin or ISA. If
you're hungry for onlyasmall meal, eat halfor aquarter ofyour usual
protein andcarbohydrate, and cover it with only halfor a quarter of
your usual dose of insulinor ISA.
Like hypoglycemia, dehydrating illness canbelife-threatening to a di
abetic. Encourage the people youlive withto readthis chapter care
fully. The supplies mentioned should be kept in locations known to
all. Phone yourphysician at thefirst sign of fever, diarrhea, or vomit
ing. The chances are that he/she would much rather be contacted
early, when dehydration andloss of blood sugar control canbe pre
vented. Emergency situations make treatment moredifficult, so you
canmake yourlife andyour physician's a bit easier byphoning before
major problems occur.
Your physician will probably ask you whether yoururine shows ke
tones, so use the Ketostix whenever you urinate beforeyou call.Also,
let your doctor know if you have taken any aspirin in the prior 24
hours, as this can cause a false positive Ketostix reading.If you are not
eating, your urine will certainly show"moderate" ketones. Your physi-
How toCope with Dehydration, DehydratingIllness, and Infection 355
cian should therefore only be concerned if it shows "high" ketones
combinedwith high blood sugars (160 mg/dl or above). Always report
yourrecent bloodsugars when you phone yourphysician.
Most infections cancause elevation of blood sugars, from an infected
toe to infected tonsils to infected heart valves. Most infections cause
symptoms that are recognizable, such as burning upon urination if
youhave a urinary tract infection, coughing if you have bronchitis,
and soon. So you'll get pretty prompt warning from your body that
you should immediatelycontact your physician. Ifyou have type 2di
abetes or early type 1, you certainly don't want your blood sugars to
get sohigh thatyour remaining beta cells are destroyed. My friend Jay
put offvisiting a urologist until his blood sugars got sohigh that his
type 2diabetes became type 1diabetes andherequired 5daily insulin
injections. Occult, or hidden, infections will not become readily ap
parent unless younotice that your blood sugars have become unrea
sonably high andyou have the good judgment tocontact your doctor.
By far the most common type of occult infection is that family of
infections that affect dental structures. This includes infections that
affect root canals, gums, andjawbones. Ahistory of elevated blood
sugars over a period of years predisposesdiabetics to such infections;
these infections, inturn, predispose diabetics tohigh blood sugars and
severe insulin resistance.
If one of my patients calls ouroffice and complains of recent onset
high blood sugars but no apparent accompanying infection (no
coughing, for instance), we ask ifshe orhe is reusing insulin syringes
and contaminating insulin, making injections relatively ineffective
(seepage 257). If the answeris no, then we recommend a visit to the
dentist immediatelyto searchfor an oral infection.
Among thethings thatyour dentist should doare toexamine your
gumsverycarefullyand to tap every tooth to seeif one or more is ten
der. Heor she should also touch each tooth with a chip of ice. Pain
uponexposure to cold is the most common overt symptom of infec
tion inthetooth or jawbone, inmy experience. We have hadpatients
with dentists who refused todothis and we've had toinstruct thepa
tients tofind better dentists. This is one ofthose many cases of being
a good, educated health care consumer in order to get proper treat-
356 Treatment
ment for your diabetes. In each case, when a newdentist performed
thesetests, a problem was found. If yourdentist does find a problem,
he or shewill probably refer youto anendodontist or periodontist to
treat the infection.
Even after such dental infections have been successfully treated,
however, bloodsugar elevations frequently continue for manymonths.
If this occurs, anappropriate antibiotic shouldbe prescribed andcon
tinued until blood sugars remain at their preinfection level. Many
people require continuation of antibiotics for as long as a year after
treatment to prevent further bloodsugar increases. When using oral
antibiotics, always take aprobiotic every day,* at least 2 hours before
or after the antibiotic, to replace gastrointestinal bacteria killedby the
To help prevent dental infections, it is wise to arrange with your
dentist for tartar to be removed from your teeth ultrasonically every
three months. You shouldalso brush your teeth at leasttwice dailyand
aftermeals floss frombetween yourteethanyfoodthat remains there.
If your teeth are tootightly spaced for flossing, try Stimudent, which
isaspecially designed toothpick withatriangular cross-section. Push
it between yourteeth withthe base of the triangle against your gum.
An eveneasier productto useis Doctor's BrushPicks, available at most
*My current favorite probiotic issaccharomyces boulardii (brand name Floras-
tor). It is availableat most pharmacies.
Delayed Stomach«Emptying:
A number of times throughout this book, you've come across
the terms "delayed stomach-emptying" and "gastroparesis."
As I explained in Chapter 2, elevated blood sugars for pro
longed periods can impair the ability ofnerves to function properly.
It's very common that the nerves that stimulate the muscular activity,
enzyme secretion, and acid production essential to digestion function
poorly in long-standing diabetes. These changes affect the stomach,
the gut, or both. Dr. Richard McCullum, a noted authority ondiges
tion, has said that ifa diabetic has any other form ofneuropathy (dry
feet, reduced feeling in the toes, diminished reflexes, et cetera), he or
she will also experience delayed or erratic digestion.
Slowed digestion can befraught with unpleasant symptoms (rarely),
or it may only be detectable when we review blood sugar profiles
(commonly) or perform certain diagnostic tests. The picture isdiffer
ent for each of us. For more than twenty-five years, I suffered from
many unpleasant symptoms myself. I eventually saw them taper off
and vanish after thirteen years of essentially normal blood sugars.
Some ofthephysical complaints possible (usually after meals) include
burning along the midline ofthe chest ("heartburn"), belching, feel
ing full after a small meal (early satiety), bloating, nausea, vomiting,
constipation, constipation alternating with diarrhea, cramps a few
inches above the bellybutton, and an acidtastein the mouth.
Most of these symptoms, as well as effects upon blood sugar, relate to
delayed stomach-emptying. This condition iscalled gastroparesis dia-
beticorum, which translates from the Latin as "weak stomach of dia
betics." It is believedthat the major causeof this condition is neuropa
thy(nerve impairment) of thevagus nerve. This nerve mediates many
of the autonomicor regulatory functions of the body, including heart
rate and digestion. In men, neuropathy of the vagus nerve can also
lead to difficulty in achieving penile erections. To understand the ef
fects of gastroparesis, refer to Figure 22-1.
Ontheleft is a representation ofa normal stomach after a meal. The
contents are emptying into the intestines, through the pylorus. The
pyloric valve is wide open (relaxed). The lower esophageal sphincter
(LES) is tightly closed, to prevent regurgitation of stomach contents.
Notshown isthegrinding andchurning activity of the muscular walls
of the normal stomach.
Onthe right is pictured a stomach with gastroparesis. The normal
rhythmic motions ofthe stomach walls are absent. The pyloric valve is
tightly closed, preventing the unloading of stomach contents. Atiny
opening aboutthesize ofa pencil point may permit asmall amount of
fluid to dribble out. When the pyloric valve isin tight spasm, some of
us cansometimes feel a sharpcrampabove the belly button. Since the
Pyloric Valve
Normal Stomach Paretic Stomach
Fig. 22-1. Normal andparetic stomachs.
Terry Eppridge
Delayed Stomach-Emptying: Gastroparesis 359
lower esophageal sphincter (LES in Figure 22-1) is relaxed or open,
acidic stomach contents can back upintotheesophagus (thetubethat
connects the throat to the stomach). This can cause aburning sensa
tionalong themidline of thechest, especially while theperson islying
down. I have seen patients whose teeth were actually eroded over time
by regurgitated stomach acid.
Because the stomach does not empty readily, one may feel full even
after a small meal. In extreme cases, severalmeals accumulate and cause
severe bloating. More commonly, however, youmayhave gastroparesis
andnot be aware of it. In mildcases, emptying maybe slowed some
what, but not enough to make youfeel any different Nevertheless, this
can cause problems withblood sugar control. Consuming certain sub
stances, such as tricyclic antidepressants, caffeine, fat, and alcohol, can
further slowstomach-emptying and other digestive processes.
Some years ago, I received a letter frommy friend Bob Anderson.
His diabetic wife, Trish, whowas not my patient andhas since passed
away, hadbeenexperiencing frequent loss of consciousness fromsevere
hypoglycemia, caused by delayed digestion. His description of anen
doscopic exam, when he was allowed to lookthrough a flexible tube
into Trish's stomach and gut, paints agraphic picture.
Allthisbrings metotoday's endoscopy exam. I watched through
the scope andfor the first time, I nowunderstand whatyouhave
been saying about diabetic gastroparesis. Not until I viewed the
inside of the duodenum didI understand the catastrophic effect
of 33 years of diabetes uponthe internal organs. There was al
most no muscle action apparent to move food out of the stom
ach. It appeared as a very relaxed smooth-sidedtube insteadof
having muscular ridges ringing thepassage. I suppose apicture is
worth a thousand words. Diabetic neuropathy is more than a
manifestation of atilting gait, blindness, andother easily observ
able presentations; it wrecks the whole system. This you well
know. I am learning.
Consider the individual who has verylittle phase I insulin release and
must take fast-acting insulin orone of theolder-type (sulfonylurea) or
360 Treatment
newer pancreas-provoking OHAs before each meal. If he were to take
hismedication andthenskip themeal, hisbloodsugarwouldplummet.
When thestomachempties tooslowly, it canhave almost thesame effect
as skipping a meal. If we knew when the stomach would empty, we
coulddelay the insulin shot or addsome NPHinsulin to the regular to
slowdownitsaction. Thebigproblemwithgastroparesis, however, isits
unpredictability. We never know when, or how fast, the stomach will
empty. If the pyloric valve is not in spasm, the stomach contents may
empty partially within minutes andtotally within 3hours. Onanother
occasion, when the valve is tightly closed, the stomach may remain
loaded fordays. Thus, bloodsugar mayplummet 1-2hours after eating,
andthenrise veryhigh, say 12 hours later, after emptying eventually oc
curs. It is this unpredictability that can make bloodsugar control im
possible ifsignificant gastroparesis isignored inpeople who take insulin
(or thetype of OHAs I don't recommend) before meals.
For most type 2 diabetics, fortunately, even symptomatic gastro
paresis may not grosslyimpede blood sugar control, because they may
still produce some phase I and phase II insulin. They therefore may
not require significant amounts of injected insulin to cover theirlow-
carbohydrate meals. Much of their insulin is producedin response to
blood sugar elevation. Thus, if the stomach does not empty, onlythe
lowbasal (fasting) levels of insulin are released, and hypoglycemia
does not occur.Of course, the sulfonylurea and similar OHAs (which
I don't recommend) can cause hypoglycemia under such circum
stances. If the stomach empties continually but veryslowly, the beta
cells of mosttype 2swill produce insulin concurrently. Sometimes the
stomach mayempty suddenly, as the pyloric valve relaxes. This will
produce a rapid blood sugar rise, caused bythesudden absorption of
carbohydrate following the entrance of stomach contents into the
small intestine. Most beta cells of type 2 patients then cannot counter
rapidly enough. Eventually, however, insulin release catches up and
blood sugar drops to normal, if a reasonable regimen is followed. If
your supper doesn't fully leave your stomach before you sleep, you
mayawaken witha highmorning bloodsugar due to emptying over
night, even though your bedtime blood sugar was lowor normal.
In anyevent, if you do not require insulin or use a sulfonylurea-
type OHAbefore meals, there isnohazard ofhypoglycemia duetode
layed stomach-emptying. This assumes thatanylong-acting insulin or
sulfonylurea is administered indoses that cover onlythe fasting state,
as discussed in prior chapters. The traditional use of large doses of
DelayedStomach-Emptying: Gastroparesis 361
these medications, meant to cover both the fasting and fed states,
brings with it the hazard ofpostprandial hypoglycemia when gastro
paresis is present.
Efforts at diagnosis are usuallyunnecessaryifthere isnoreason tosus
pect the presence of gastroparesis. So first we must havean index of
suspicion. If, at theinitial history-taking interview with your physi
cian, you mention symptoms like those described earlier inthis chap
ter, he should have a high index of suspicion. If your R-R interval
study (Chapter 2) at the initial physical exam isgrossly abnormal, he
can bequite certain ofgastroparesis. Remember thatthis studychecks
theability of thevagus nerve to regulate heart rate. If thenerve fibers
goingto theheart areimpaired, thebranches that activate the stomach
are probablyalso impaired. Inmy experience, the correlation ofgrossly
abnormal R-R studies with demonstrable gastroparesis isvery real.*
Diagnostic Tests
Given the physical symptoms or the abnormal R-R study, your physi
cian may want to consider further tests to evaluate your condition.
The most sophisticated of these studies isthegamma-ray technetium
scan. This test is performed at many medical centers, and is quite
costly. It works this way: You eat some scrambled eggs to which a
minute amountof radioactive technetium has been added. Agamma-
ray camera trained on your abdomen measures (from outside your
body) thelowlevels ofradiation thetechnetium emits astheeggs pass
from your stomach intoyour small intestine. If thegamma radiation
drops off rapidly, the studyis considered normal.
Aless precise study can beperformed at much lower cost byany ra
diologist. This iscalled thebariumhamburger test. Inthistest, youeat
a V4-pound hamburger andthendrinkaliquid that contains theheavy
element barium. Every halfhourorso, anX-ray photo istaken ofyour
stomach. Since the barium shows up in these photos, the radiologist
*If, during anR-Rstudy, your heart rate varies only 28 percent between inhaling
and exhaling, then you will likely have mild gastroparesis. If the variation is
about20percent, gastroparesis will probably bewhatI call moderate, andif less
than 15percent, I would call it severe.
362 Treatment
can estimate what percent of the bariumremains in your stomachat
the end of each timeperiod. Total emptying within3 hours or less is
usually considered normal.
Despite their theoretical usefulness, neither of these studies is any
where near 100 percent sensitive, because of theunpredictable nature
oftheparetic stomach. One day it may empty normally, another day it
maybeabitslow, and onyet another day its emptying maybeseverely
delayed. Because of this unpredictability factor, thestudy may have to
berepeated anumber oftimes before adiagnosis can bemade. Thepos
sibility exists that you could have several normal studies but still have
abnormal stomach-emptying. I therefore advise my patients against
using either ofthese two tests. The R-R study ismy gold standard.
Telltale Blood Sugar Patterns
Having medical tests is badenough, but having to repeat themwith
conflicting results naturally proved quite annoying to my patients
manyyears agowhenI actually repeated them.Worse than annoyance,
the studies are not cheap, and most insurance companies will not
pay for repeats of the same study unless they're separated by many
months. If you're regularly measuring yourbloodsugarlevels and try
ing to keep them in the normal range, it's really not difficult to spot
gastroparesis that'ssevere enough to affect bloodsugars. Forpractical
purposes, thisisjustthedegree ofgastroparesis thatshould concern us.
Below are some of the typical blood sugar patterns that I look for.
To call these patterns, though, is slightly misleading. The hallmark of
gastroparesis is randomness, unpredictability from one day to the next.
These "patterns" come andgoin such a fashion that bloodsugarpro
files arerarely similar on 2or 3 successive days. Thefirst twopatterns
together arehighly indicative of gastroparesis, while thethirdbyitself
is usually adequate for diagnosis.
• Lowblood sugar occurring1-3 hours after meals.
• Elevated bloodsugar occurring 5or morehours aftermeals with
no other apparent explanation.
• Significantly higher fasting bloodsugars in the morningthan at
bedtime, especially if supperwas finished at least 5 hours before
retiring. If bedtime long-acting insulin or ISA is gradually in
creased in an effort to lower the fasting blood sugars, we may
find that the bedtime dose is much higher than the morning
dose. On some days fasting blood sugar maystill be high, but on
DelayedStomach-Emptying. Gastroparesis 363
other days it maybe normal or even too low. We're thus giving
extra bedtime medicationsto accommodate overnight stomach-
emptying—but sometimes the stomachdoesn'temptyovernight
and fastingblood sugars drop too low.
Having seen such patterns of blood sugar, we can then perform a
simple experiment to confirm that they reallyare caused by delayed
Skipsupper and its premeal insulinor ISA one night. When you go
to bed, be sure to take your basal (bedtime) insulin or ISA, measure
your blood sugar,and then measureyour fasting blood sugar the next
morning on arising. If,without supper,your blood sugar has dropped
or remained unchanged overnight, gastroparesis is the most likely
cause of the roller-coaster morning blood sugars.
Repeat this experiment severaldayslater, and again a third time, af
ter another fewdays. If each experiment results in the same effect,de
layed stomach-emptying is virtually certain on one or more of the
nights whenyouhad eaten.Whenyouhad previouslybeen eatingsup
pers, at least some of the following mornings had shown an overnight
rise in blood sugars. Sincesuch risesoccurred on nights when you had
eaten supper, but noton the nights when you did noteat, the rise must
have been caused by food that did not leave your stomach until after
you went to bed. Be very cautious when performing this experiment,
as you may experience severe hypoglycemia upon arising or during
the night. To play it safe, checkyour blood sugar midway through the
night and correct it if it's belowyour target.
"False Gastroparesis"
I've seen a number of patients whose blood sugar profile or physical
symptoms could have been diagnostic of gastroparesis, yet their R-R
interval studies were normal or only slighdyimpaired. These people
had delayed stomach-emptying but well-functioning vagus nerves.
The conflictingdata obligedme to order upper gastrointestinal endo
scopicstudies for these people. Endoscopyusesa thin, flexible, lighted
fiber-opticcableto look directlyinto the stomach and duodenum.
The endoscopic tests demonstrated that they all had abnormalities
unrelated to their diabetes. Such findings have included gastric or
duodenal ulcers, erosive gastritis, irritable gastrointestinal tract, hiatal
hernia, and other gastrointestinal disorders such as tonic or spastic
stomach. Each of these conditions required treatment distinct from
364 Treatment
treatment for diabetes. Only with hiatal hernias were we unable to at
least partially alleviate the digestive problem. In such cases, however,
surgical correction of the hiatal hernia is possible, but it may or may
not normalize emptying. Blood tests for parietal cell antibodies and
serumvitamin B-12might be performed to rule out autoimmune gas-
tropathy as a causeof gastritis.
The following suggestions for treating gastroparesis may or may
not facilitate stomach-emptyingfor the aboveconditions but should
certainly be tried. The loud and clear message from this is that the
R-R interval study should be performed on every diabetic patient
whoseblood sugar profiles resemble those outlined above.
It is worth noting that gastroparesis can be curedby extendedperiods
of normal blood sugars. I've seen several relatively mild cases where
special treatment was terminated after about 1year, and blood sugar
profiles remained flat thereafter. At the same time, R-R studies im
proved or normalized. Sincemy late teens, I experienced severedaily
belching, and burning in my chest. These symptoms gradually eased
off, and eventually disappeared, but onlyafter thirteen years of nearly
normal blood sugars. Mylast R-Rstudy was normal. The "sacrifices"
in lifestyle required for treatment of gastroparesis may reallypay off
months or years later. The vagusnerve doesn't control only stomach-
emptying — there are a number of other complicationsresulting from
impairedvagus function that can be reversed by maintainingnormal
bloodsugars. Theregained ability to sustaina penileerectionisan im
portant one for many of my male patients.
Once gastroparesis has been confirmed as the major cause of high
overnight blood sugars and wide random variations in blood sugar
profiles, we can begin to attempt to control or minimize its effects. If
your blood sugar profiles reflect significant gastroparesis, there is no
way to get them under control by only juggling doses of insulin.
There's just too much danger of either very high or very low blood
sugars for such approaches to work. The only chance for effective
treatment is to concentrate on improving stomach-emptying.
How do we do this?
We have four basic approaches. First is the use of medications. Sec-
Delayed Stomach-Emptying: Gastroparesis 365
ond is special exercises or massage during and after meals. Third is
meal plan modification utilizing ordinary foods, and fourth is meal
plan modification utilizingsemiliquid or liquid meals.
It's unusual for a singleapproach to normalize blood sugar profiles
fully, so most often we try a combination of these four approaches,
adapted to the preferences and needs of the individual. As these at
tempts start to smooth out blood sugars,we must modify our doses of
insulin or ISAs accordingly. The guidelinesthat we use to judge the ef
ficacyof a given approach or combination of approaches are these:
• Reduction or elimination of physical complaints such as early
satiety,nausea, regurgitation, bloating, heartburn, belching, and
• Elimination of random postprandial hypoglycemia
• Eliminationof random, unexpectedhigh fastingblood sugars—
probablythe most commonsignof gastroparesis that weencounter
• Flattening out of blood sugar profiles
Remember that the last three of these improvements may not be
possible even without gastroparesis if you're following conventional
dietary and medication regimens for "control" of your blood sugar.
For example, I knowof no waythat will truly flatten out blood sugar
profilesif you're on a high-carbohydrate diet and the associatedlarge
doses of insulin.
Medications That Facilitate Stomach-Emptying
There is no medication that will cure gastroparesis. The only "cure" is
months or years of normal blood sugars. There are, however, some
pharmaceutical preparations that may speed the emptying of your
stomach after a meal if your gastroparesis is only mild or moderate in
severity (page 361, footnote). These will help smooth out your blood
sugar profiles after that meal. Most diabetics with mild to moderate
gastroparesiswill require medication before everymeal.
When gastroparesisis verymild, it maybe possibleto get awaywith
medication only before supper. For some reason — perhaps because
most people tend not to be as physically activeafter supper, and may
havetheir largest meal of the dayin the evening— digestionof supper
appears to be more impaired than that of other meals. It is also likely
that stomach-emptying is slowerin the evening, evenfor nondiabetics.
Medications for gastroparesismay take the form of liquids or pills.
The question immediatelyarisesthat if pillsmust dissolve in the stom-
366 Treatment
ach to become effective, just howeffective are they going to be?My ex
perienceis that they're of questionable valueunlesschewed. The time
required for a pill to dissolve in a paretic stomach is likely to be
lengthy, and consequently the medicationmay take several hours to
become effective. I generally prescribe only liquid medications or
chewedtablets for stimulating gastric (stomach) emptying.
Cisapride suspension (Propulsid, Janssen Pharmaceutica) stimu
lates the vagus nerve to facilitate stomach-emptying. I usually pre
scribe 1 tablespoon (25 mg), 15-30 minutes before meals for adults.
Many people will require 2 tablespoons for maximum effect. Larger
doses appear to be of little added value. The manufacturer recom
mends doses only up to 20 mg (2 teaspoons) for the treatment of
esophageal refluxdisease. This condition is much more responsive to
treatment than is diabetic gastroparesis, which as a rule requires the
larger doses. The package insert also recommends a bedtime dose,
which serves no purpose for gastroparesis. In many cases, cisapride
alone will not bring about completestomach-emptying.Wemay add
other medicationsif blood sugar profiles don't level off.
Cisapride can inhibit or compete for liver enzymes that clear cer
tain medications from the bloodstream. Your physician should there
fore reviewall your medications, especiallyantidepressants, antibiotics,
and antifungal agents, before prescribing cisapride. Stimulating the
vagus nerve will alsoslowthe heart. Sincediabeticswith gastroparesis
usually have an excessively rapid heart rate (more than 80 beats per
minute) this is not often a problem. Some individuals, however, have
a cardiac conduction defect that abnormally slowsthe heart. For such
people, cisapridecanstop the heart, resultingin death. Since, for many
years, physicians have ignored this bold warning on the packagein
sert, a number of deaths actually have occurred. The product has
therefore been removed from the marketplace in many countries. It is
still available in the United States at no charge as an "investigational
drug" if prescribed by a gastroenterologist who has been cleared for
its use by the investigational review board of his hospital. It is also
available from pharmacies in New Zealand under the trade name
Prepulsid (not a misspelling, a different brand name for the same
product). It maybe purchasedvia the Internet after searchingfor Web
sitescontainingthe words"pharmacy"and"NewZealand." Sucha pur
chase will not be covered by your insurance (unless you happen to
Delayed Stomach-Emptying. Gastroparesis 367
live in New Zealand), and withshipping charges, it canbe expensive.
Furthermore, of late, manypharmacies in New Zealand refuse to ship
to the United States. Since this agent works by stimulating the vagus
nerve, it will not produce results if the nerve is almost dead — as with
heart rate variability less than 13percenton an R-R study.
Super Papaya Enzyme Plus has been praisedby many of my pa
tients for its rapid relief of some of the physical symptoms of gastro
paresis— bloating and belching, for example. Someclaimthat it also
helps to level off the blood sugarswings caused by gastroparesis. The
product consists of pleasant-tasting chewable tablets that contain a
varietyof enzymes (papain,amylase, proteases, bromelain,Upase, and
cellulase) that digest protein, fat, carbohydrate, and fiber while they
arestillinyourstomach. You would normally chew3-5 tablets during
and at the end of each meal. The tablets are available in most health
food stores and are marketed by American Health, (631) 567-9500,
Ronkonkoma, NY11779. Theyare alsoavailable fromRosedale Phar
macy, (888) 796-3348. Someof mykosherpatients usea similar prod
uct called Freeda All Natural Parvenzyme, which is distributed by
FreedaVitamins,36EastForty-firstStreet,NewYork, NY10017, (800)
777-3737, and on the Web at freedavitamins.com. The small amount
of sorbitoland similar sweeteners contained in these productsshould
not have a significant effect on your bloodsugar if consumption is
limited to the above dose.
Domperidone (Motilium, Janssen Pharmaceutica) is not yet avail
ablein the UnitedStates. It canbe purchasedin Canada, the U.K., and
perhaps some other countries. Pharmacies in Canada are no longer
permittedto shipmedications to theUnited States unless theyarepre
scribed byaCanadian physician. It therefore may benecessary to pur
chase it elsewhere viatheInternet.* Since it isnot available asa liquid,
weaskpatientsto chew2tablets (10mgeach) 1hour beforemeals and
to swallow with 8 ounces of water or diet soda. I limit dosing to 2
tablets because largerdoses cancause sexual dysfunction in men and
absence of menses in women. These problems resolve whenthe drug
*Anumber of Canadian pharmacies for an additional charge of $5 cansecure
prescriptions for distant foreign patientsfromCanadianphysicians. One of these
is MurrayShore Pharmacy, (800) 201-8590.
368 Treatment
is discontinued. Since it works by a mechanism different from those of
the preceding products, its effects can be additive (that is, useful with
other preparations). Janssen maymarket a liquid form of this product
in the United States at some time in the future. In the meantime, some
gastroenterologists maybe ableto prescribe it, likePropulsid,as an in
vestigational drug.
Metoclopramide syrup maypossibly be the most powerfulstimu
lant of gastricemptying.It worksin a fashionsimilarto domperidone,
by inhibiting the effects of dopamine in the stomach. Because it can
readily enter the brain, it can cause serious side effects, such as som
nolence,depression, agitation,and neurologicproblemsthat resemble
Parkinsonism. These side effects can appear immediatelyin some in
dividuals or only after many months of continuous use in others. Be
cause gastroparesis often requires doses high enough to cause side
effects, I usethis medicationinfrequently and limit dosingto no more
than 2 teaspoons 30 minutes before meals.
If you use metoclopramide, you should keep on hand the antidote
to its sideeffects — diphenhydramine elixir (Benadrylsyrup). Twota
blespoons usually work.Ifside effects become serious enough towarrant
useof theantidote, themetoclopramide should beimmediately andper
Abrupt discontinuation of metoclopramide has been reported to
cause psychotic behavior in two patients after continuous use for
more than three months. This information might suggest to your
physicianthat it be graduallytapered off if it is to be discontinued af
ter even two months of continuous use.
Erythromycin ethylsuccinate is an antibiotic that has been used to
treat infections for manyyears. It has a chemical composition that re
sembles the hormone motolin, which stimulates muscular activity in
the stomach. Apparently, when stimulation of the stomach by the va
gus nerve is depressed, as with autonomic neuropathy, motolin secre
tion is diminished. Three papers deliveredto the 1989annual meeting
of the American Gastroenterological Association demonstrated that
this drug can stimulate gastricemptying in patients with gastropare
sis. In people without gastroparesis, erythromycin can cause nausea,
unless taken after drinking fluids. I ask my patients to drink two
glasses of water or other fluid before each dose. I prescribe erythro-
DelayedStomach-Emptying: Gastroparesis 369
mycinethylsuccinate oral suspension just beforemeals. Westart with
1 teaspoon of the 400 mg/tsp concentration, and increase to several
teaspoons if necessary. As each teaspoon of this suspension contains
3.5 grams of sucrose (table sugar), it will be necessary to increase
slightly the dosesof insulin covering mealsto reduceblood sugar ele
vation whilethis medication is used. If the liquid is kept in a refriger
ator, the tastebeginsto deteriorateafter 35days. At room temperature,
taste deteriorates after 14 days. I have seen no side effects from this
medication. I insist that patients who use it chronically take 1 probi
otic capsule (such as Florastor [saccharomyces boulardii], Culturelle
Lactobacillus GG, or Nature's Way Primadophilus Reuteri) at least 2
hours before or after each dose. This is to restore to the intestine nat
ural bacteria that can be destroyed by this antibiotic. It is also wiseto
consume one 150mg fluconazole tablet per month to inhibit growth
of fungus in the GI tract or vagina. I have not found erythromycin
to be especially effective for treatinggastroparesis, despite published
Betaine hydrochloride with pepsin is a potent combinationthat
can predigest food in the stomachby increasing acidityand adding a
powerful digestive enzyme. It can be procured at most health food
stores or at Rosedale Pharmacy. Because of its acidityit should not be
usedbythose with gastritis, esophagitis, or stomach/duodenal ulcers.
Food that has been predigestedwill more likely pass through the nar
rowedpyloricvalve of gastroparesis. We initially use 1tablet or capsule
midmeal. If no burning is perceived, we increase the dose to 2 and
then eventually3 tablets or capsulesspacedevenlythroughout subse
quent meals. It should never bechewed or taken onan empty stomach.
Sincebetaine HC1 with pepsin, unlike cisapride, does not attempt to
stimulate the vagusnerve, it is frequentlyof valuefor evenseverecases
of gastroparesis.
Nitric Oxide Agonists
Although the aforementioned agents can be veryeffective when gas
troparesis is mild, their effectiveness in minimizing blood sugar un
certainty after meals diminishes when this condition is more severe.
Myfrustration in trying to circumvent this problemhas led to my in
vestigation of a class of substances called nitric oxide agonists. Such
agents are currently being used to relieve effects of angina in patients
370 Treatment
with cardiac disease. Since they workby relaxing the smooth muscle in
the walls of coronary arteries, I assumedthat they could also relax the
smooth muscle of the pyloricvalve.
My initial trial was with a medicationcalled isosorbide dinitrate. I
hadit prepared asasuspension in almondoil (with flavoring) sothat it
couldcoatthe pylorus andwork directiy upon it. I hadit compounded
in aconcentrate of 5mg/tsp (1 mg/ml). I waspleasedto seethat my as
sumption proved correct — it was veryeffective fornearlyallofmy pa
tients who used it. Thus far, it appears to be more successful than any
of the agents described above. Nevertheless, it is only partially effective
for more severe cases of gastroparesis.
This formulation can be prepared by any compounding chemist
(see footnote, page 202). The only adverse effect I've observed has
been headache in about 10 percent of the users. Although the head
ache usually resolves after several days of use, I try to prevent it by
startingwith very smalldoses that canthen be gradually increased.
I therefore recommend that initially lh teaspoon be taken 30-60
minutes before dinner. After one week, we increase the dose to 1 tea
spoon. If this fails to level off blood sugars at bedtime and the follow
ing morning, we continue 1teaspoon for aweek and then increase it
to 2 teaspoons. If this is not fully effective, we then increase to 3 tea
spoons. If this dose doesn't do the trick, I discontinue the treatment, as
further increases areunlikely to be effective. If 1-3 teaspoonswork, we
then use the same dose 30-60 minutes before each meal. It's been un
usual forthis formula to be totallyineffective. The liquid must be vig
orously shaken before use.
If you have a cardiac condition, isosorbide dinitrate should not be
used for gastroparesis unless approvedby your cardiologist.
Unfortunately, isosorbide dinitrate usually stops working after a
period of weeks to months. I therefore attempt to increase effective
ness and lower blood sugar levels by applying a chemically similar
product to the skin direcdy overthe pylorus. What I prescribe is ani
troglycerine skin patch. These are available by prescription at any
pharmacy in strengths of 0.1,0.2,0.4, and 0.8 mg. The patchis placed
over the pylorus, which is located on the midline of the abdomen
above the navel, about 1V4 inches (37 mm) below the middle of the
lowest ribwhere it forms aninvertedV. The patchis appliedon arising
in the morning and removed at bedtime. We start with the 0.1 mg
patch and increase the sizeeachweek if there areno adverse effects. As
with isosorbide dinitrate, nitroglycerine should not be used for gas-
Delayed Stomach-Emptying. Gastroparesis 371
troparesis without your cardiologist's approval if you have a cardiac
Anotheralternative isthe clonidine adhesive skinpatch.Thisprod
uct is soldas Catapres in all pharmacies to lower bloodpressure and
requires a prescription.It isa powerful smooth musclerelaxant.It can,
however, cause somnolence (sleepiness) in some people. We therefore
start at the smallest size (1 mg) for the first weekand increaseit to 2
mg for the secondweek, then 3 mg for the third weekand thereafter.
Althougheachpatch willworkfor a week on most people,we remove
it at bedtime and replace it the next morning. Since the patch's adhe
siveness will be reduced after it's removed, you can use paper tape to
keep it attached after the first day. If it causes tiredness, we lower the
patch dosageor discontinue it.
Like the aforementionednitric oxide agonists, it can stop working
eventually. If it has been effective and stops working, we discontinue it
and restart it after a couple of months. Somepatients find that a patch
will stop working after 3-4 days. For these people, we change to a new
patch midweek.
The reasonwe removethe clonidine(or nitroglycerine) patch from
the skin at bedtime isto slowdownthe developmentof toleranceto its
actionwhicheventuallyoccurs. I alsorecommendalternatingdaytime
skin patches— one weekon clonidineand one week on nitroglycer
ine — alternating over and over.
Exercises That Facilitate Stomach-Emptying
The pareticstomachmaybe described asa flaccid bag,deprivedof the
rhythmic muscular squeezingpresent in a stomach that has a properly
functioning vagus nerve. Any activity that rhythmically compresses
the stomach can crudely replicate normal action. You may perhaps
have observed how a brisk walk can relieve that bloated feeling. I
thereforestronglyrecommendbriskwalking for an hour immediately
after meals — especiallyafter supper.
A patient of mine learned a trick from her yoga instructor that
eliminated the erratic blood sugar swings caused by her moderate
gastroparesis. The trick is to pull in your belly as far as you can,
then push it out all the way. Repeat this with a regular rhythm as
manytimes as you can, immediately after eachmeal. Over a period of
weeks or months, your abdominal muscles will become stronger and
stronger, permitting progressively more repetitions before you tire.
Eventually shoot for several hundred repetitions — the more the bet-
372 Treatment
ter. This should require lessthan 4 minutes of your time per hundred
reps, a small price to pay for an improvement in your blood sugar
Another patient discovered an exercise that I callthe "backflex." Sit
or stand while bending backward as far as you can. Then bend for
ward, about the same amount. Repeat this as many times as you can
Although these exercises may sound excessively simple, even silly,
they havehelpedsomepeoplewith gastroparesis.
Mechanical Aids
Hand-held massager. One product of possible value is a variable-
speed hand-held massager that can be placed over the stomach (left
side of the abdomen just below your ribs). A 15-30 minute massage
mightspeedstomach-emptying. Thisproduct iscalled Programmable
Percussion Massager withHeat#HF755 and isavailable fromSharper
Image Corp., (415) 445-6000 or online at www.sharperimage.com.
Usethe largestof the five setsof removable heads.
Chewing Gum Can Make a Big Difference
Theact of chewing produces saliva, which not onlycontains digestive
enzymes but also stimulates muscular activity in the stomach and
tends to relax the pylorus. Orbit is a delicious "sugarless" gumwith a
long-lasting flavor. It contains only1gramof sugar per piece and so
will have littleeffect uponyourbloodsugar.* Chewing gumfor at least
1hour aftermeals isaveryeffective treatmentof gastroparesis outside
of major dietarychanges. Don't chewone piece afteranother, because
the grams of sugar can add up.
Meal Plan Modifications, Utilizing Ordinary Foods
Moreoften than not, changes inyourmeal planwill prove moreeffec
tive thanmedication. The problem isthatsuch changes are unaccept
able to many patients. We usually proceed from most to least
convenient in six stages:
*Another worthwhile chewing gumisXlearDent Each piece contains %gramof
xylitol, which can kill the bacteria that cause tooth decay. To order it, phone
(877) 599-5327, or visit the Web at www.xlear.com.
Delayed Stomach-Emptying: Gastroparesis 373
1. Drinking at leasttwo 8-ounce glasses of sugar-free, caffeine-free
fluid while eating,and chewingslowlyand thoroughly
2. Reduction of dietary fiber or first running fiber foods through a
blender until nearlyliquid.
3. Virtual elimination of unground red meat, veal, pork, and fowl
4. Reduction of protein at supper
5. Introduction of four or more small dailymeals, insteadof three
larger meals
6. Semiliquid or liquid meals
In the pareticstomach, soluble fiber (gums) and insoluble fiber can
form a plugat the very narrowpyloricvalve. This is no problem forthe
normal stomach, wherethe pyloric valve is wide open. Many patients
with mild gastroparesis have reportedbetter relief of fullness and im
provedblood sugar profiles after modifyingtheir diets to reduce fiber
content or to renderthe fiber more digestible. This means, for example,
that mashedwell-cookedvegetables must be substituted forsalads, and
high-fiber laxatives such as those containing psyllium (e.g., Metamu-
cil) should be avoided. Acceptable vegetables might include avocado,
summer squash, zucchini, or mashed pumpkin (sweetened, if you like,
with steviaand flavored with cinnamon). It also meansthat you would
have to give up one of our alternatives to toast at breakfast — bran
crackers. You might want to try cheese puffs (page 178) instead.
Most people in the United Stateslike to eat their largest meal in the
evening. Furthermore, they usually consume their largest portion of
meat or other protein food at this time. These habits make control of
fasting blood sugars very difficult for people with gastroparesis. Ap
parently animal protein, especially red meat, like fiber, tends to plug
up the pylorus if it's in spasm. An easy solution is to move most of
your animal protein fromsupperto breakfast and lunch. Many of my
patients have observed remarkable improvements when they do this.
We usually suggest a limit of 2 ounces of animal protein, restricted to
fish, ground meat, cheese, or eggs, at supper.This is not very much. Yet
people are usually so pleased with the results that they will continue
with such a regimen indefinitely (of course, as protein is shifted from
one meal to another, doses of premeal insulin or ISA must also be
shifted). With a reduction of delayed overnight stomach-emptying,
the bedtime dose of longer-acting insulin or ISA may have to be re
duced so that fasting blood sugar will not drop too low.
374 Treatment
Some people find that by movingprotein to earlier meals, they in
creasethe unpredictability of blood sugar after these meals. For such a
situation, we suggest, for those who do not use insulin, four or more
smaller meals each day, instead of three larger meals. We try to keep
these meals spacedabout 4 hours apart, so that digestion and doses of
ISAfor one meal are lesslikelyto overlapthose for the next meal. This
can be impractical for those who take preprandial insulin. Remember,
you must wait 5 hours after your last shot of preprandial insulin be
fore correcting elevated blood sugars.
Both alcohol and caffeine consumption can slowgastricemptying,
as can mint and chocolate. Theseshould thereforebe avoided, espe
cially at supper, if gastroparesis is moderate or severe.
Semiliquid or Liquid Meals
Alast resort for gastroparesis is the use of semiliquidor liquid meals. I
say"last resort"becausesucha restrictiontakesmuch of the pleasureout
of eating,but it maybe the onlywayto assure near-normal blood sug
ars.Withthisdegree of bloodsugarimprovement, the gastroparesis may
slowly reverse, as mine did. The restriction can then eventually be re
moved. In thissection I'll try to give yousomeideas that youcanuseto
createmeal plans usingsemiliquid foodsthat still follow our guidelines.
Baby food. Low-carbohydrate vegetables and nearly zero carbohy
drate meat, chicken, and eggyolkprotein meals are readilyavailableas
babyfood. Rememberto readthe labels. Also remember that for a typ
ical protein food, 6 grams of protein on the label corresponds to about
1 ounce of the food itself by weight. To avoid protein malnutrition,
youshouldconsume at least 1gramof proteinfor everykilogram (2.2
pounds) of ideal body weight. Thus, a slim person weighing 150
pounds (68 kilograms) should consume at least 68 grams of protein
daily. This worksout to about 11 ouncesof protein foods. Peoplewho
are still growing or who exercise vigorouslymust consume consider
ablymore than 1gramper kilogram of idealbodyweight.
When vegetables that only slowly raise blood sugar are ground or
mashed, they can raiseblood sugar more rapidly. So howcan we jus
tify using baby foods? The answer is that we recommend such foods
only for people whosestomachalreadyempties veryslowly. Thus even
with babyfoodyour blood sugar maystill havedifficultykeepingpace
with injected regular insulin. Later in this chapter I will show you
some tricks for circumventingthis problem.
Delayed Stomach-Emptying Gastroparesis 375
Below is a brief fist of sometypical babyfoods that can be worked
into the meal-planning guidelines set forth in Chapters10and 11. Do
not exceed those guidelines for carbohydrate, sincemost of the Laws
of Small Numbers stillapply, even if youhave gastroparesis.
Vegetables Carbohydrate
Beech Nut Green Beans (4.5-ounce jar) 8grams
Beech Nut GardenVegetables (4.5-ounce jar) 11
Heinz Squash (4.5-ouncejar) 8
Meats —Strained Protein
Beef(3-ounce jar) 2.25ounces
Chicken (3-ounce jar) 2.25
Ham (3-ounce jar) 2.25
Egg Yolks (3-ouncejar) 1.50 (plus 1gram
Unflavored whole-milk yogurt. Some brands of whole-milk yo
gurt, such as Erivan, Brown Cow Farm, or Stonyfield Farm, have no
added sugars or fruits. As noted previously, Erivan is sold at health
food stores and the other two at supermarketsthroughout the United
States. Again, always specify"wholemilk, unflavored." Remember that
"low-fat" dairy foods usually contain more carbohydrate than the
Erivanyogurt contains 11 grams of carbohydrateand 2 ounces pro
tein per 8-ounce container. Stonyfield Farm and Brown Cow Farm
both contain 12grams carbohydrate and 1.5ounces protein.
Blandfoods likeplain yogurt can be made quite tastyby adding one
of the baking flavor extracts, the powder fromtruly sugar-freegelatin
desserts (i.e., without maltodextrin), Da Vinci sugar-free syrups, or
steviawith cinnamon. The amounts used should suit your taste.
Whole-milk ricotta cheese. Whilenot as Uquid as yogurt or baby
food, ricotta cheese goes down better than solid foods. It can also be
put into a blender with some water or cream to render it more liquid.
Each 8-ounce serving of ricotta contains about 8 grams of carbohy
drate and 2 ounces protein. To my taste, ricotta is a very bland food,
but when flavored with cinnamon and stevia, it can be a real treat — a
meal that tastes like a dessert.
376 Treatment
Liquid meals. When semiliquidmeals are not fullysuccessful, the
last resort is high-protein, low-carbohydrate Uquid meals. These are
sold in health food stores for use by bodybuilders. Only use those
made from eggwhite proteins or whey, if you wish to be assured of aU
the essential amino acids. Similar products made from soy protein
may or may not contain these in adequate amounts. Many may con
tain sterols similar to estrogen.
Possible Last Resorts for Treating Gastroparesis
One of mypatients claimsthat a cosdynewtreatment has helped con
siderablyboth her gastroparesis and her neuropathic pain. It involves
the appUcation of small electriccurrents to acupuncture points on her
limbs and is caUed STS therapy. The electronic device is designated
model STS and is manufactured by Dynatronics of Salt Lake City,
(800) 432-2924. The instrument costs about $4,000 and the treatment
must be performed for 45 minutes everyday. Its effects begin after
about 2 months, and it mayactually faciUtate the heaUng of damaged
nerves. This device should not be used near an insuUn pump or by
people with implanted electrical devices.
Another costly option is electrical gastric stimulation.This involves
surgical implantation through the skin of two electrodes that contact
the muscular waU of the stomach.The connectingwiresenter a con
trol box that can be kept in a pocket or on a belt. The control unit can
be set to stimulate the stomach muscles after each meal.
Apatient from Indiana with a hiatal hernia once told me, "These Dex-
trotabs don't raise mybloodsugar one bit. What really works is one
stickof that sugar-free chewing gum" (because chewing the gum en
courages the stomachto emptya meal that maybe sittingthere).
Her comment iUustrates a majorhazardassociated with anycondi
tion that retards stomach-emptying (gastroparesis, ulcers, and so on):
treatinghypoglycemia rapidlyis nearlyimpossible. Note the qualifier,
"nearly." There are sometricksto circumvent the problem.
If your hypoglycemia occurred because your last mealis still sitting
Delayed Stomach-Emptying: Gastroparesis 377
in your stomach, you might thereafter try some chewing gum to help
it empty.
Since chewed glucose tablets can take several hours to leave your
stomach, you should suck them or, preferably, try a Uquidglucoseso
lution. Such a product is available as glucose tolerance test beverage
under a number of brand names in the United States. These include
Glucola, Limeondex, Dexicola, and Sun-Dex. The drinks are bottled
by manufacturers of clinical laboratory reagents and are stocked in
every hospital lab and private cUnical laboratory. A very convenient
version, calledGlutol, comes in plasticbottles. It is made by Paddock
Laboratories, (800) 328-5113, and is alsosoldby Rosedale Pharmacy.
It contains 2.8grams of glucose per teaspoonand 8.3grams per table
spoon. SeeTable 20-1 to calculatehow much these amounts wiU raise
your blood sugar.If you don't havea medicinespoon handy and are in
a hurry, assumethat one swaUowfromthe bottle is equivalentto 1ta
If you'retraveling and forget to bringalonga bottle of your glucose
tolerance test beverage, get some lactose-free milk. This product has
been treated with an enzyme that converts the lactose to glucose. In
the UnitedStates, the most widely marketedbrand is Lactaid. Every 4
ounces contains 6 grams of glucose. Remember, however, that Lactaid
wiU spoil after a fewdays if not refrigerated.
Even if you've used the glucose tolerancetest drink or Lactaid, you
can speed up the action by chewing gum, by doing the back-flex and
stomach exercises describedearlier in this chapter, by using a hand
held massager, and/or, prior to eating, by usingsome of the medica
tions mentioned earlier.
It takes awhile for your physician to select and fine-tune a programto
improve stomach-emptying. In the meantime, it's possible to reduce
the frequency and severity of postprandial hypoglycemia. Todo this,
you must slowthe action of preprandial insulin or ISAto match more
closely the delayyouexperience in digestingyour meals. Let's suppose,
for example, that you'U be using preprandial rosiglitazone. If youhave
378 Treatment
gastroparesis, your doctor may ask you to take it 10,30, or 45 minutes
before eating,insteadof the usual60-120 minutes. If you'll be getting
preprandial shots of regular insulin, your physician may want you to
inject immediately before eating, instead of the usual 45 minutes. If
regular stiU works too rapidly for your slow digestion, you may be
askedto take it afteryour meal.Alternatively, you might substitute 1or
more units of NPH insuUn for 1 or more units of regular in your sy
ringe, to slow the action. If, for example, you are asked to inject a
preprandial mixture containing 4 units of regular and 1unit of NPH,
you would drawthe 4 units of regular into the syringe in the usual
manner (see pages 255-256). Now insert the needle into the vial of
NPH and shakethe vial and syringetogethervigorouslya fewtimes, as
illustrated in Figure 16-6. Immediately but carefully draw 1 unit of
NPH into the syringe. Now remove the needle from the vial and draw
in about 5 units of air. The exact amount of airis not important. The
air bubble will act a bit like the metal ball in a can of spray paint to
help mix the insulins. Invert the syringe a fewtimes to permit the air
bubble to move back and forth, therebymixing the two insuUns. (This
is the only situation in which it is acceptable to mix two different in
suUns in the same syringe.)
Now you can inject the contents of the syringe, including the air.
The air wiUdissolve in your tissue fluids and cannot do any harm.
If this process confusesyou, don't worry.Your physicianor diabetes
educator should demonstrate it for you and check your technique.
If you use this procedure to slow down your preprandial dose
of regular insulin, it'U keep working for an unknown period of time
weU beyond the usual 5 hours. If you routinely correct elevatedblood
sugars with additional shots of Uspro as described on page 301, you
now have a real problem. When do you correct an elevated blood
The answer is actuaUy simple. Under these conditions, if you add
the NPH to regular beforeevery meal, you are Umited to correcting a
high blood sugar only once daily—when you arise in the morning.
This wiU be about 12hours afteryour suppertime shot of the regular-
NPH mixture. Twelve hours is more than enough time for the mixture
to have finished acting.
If you only use the NPH mixture beforedinner, then you may safely
continue to correct elevatedblood sugars before breakfast and lunch
(after waiting the usual 5 hours or more).
DelayedStomach-Emptying Gastroparesis 379
Do not use lispro to cover meals if you have delayed stomach-
emptying. The reasoning here should be self-evident. Feel free, how
ever, to use it to bring down an elevated blood sugar using the
methods previously mentioned.
Remember the insuUn-mimetic antioxidants alpha lipoic acid (ALA)
and eveningprimrose oil (EPO)? WeU, studies in the United Statesand
Germanyhaveshown them to heal the nervesinvolved in painful dia
betic neuropathy of the feet. Thesestudies achieved their results in a
matter of months, without anyattempt to control blood sugars. More
recent brief studies haveactuaUy brought about partial healing of the
vagus nerve. The studies that I read, however, utilized very high doses
of one of these agents (25,000 mg of alpha lipoic acid), administered
intravenously.Afewnaturopathic physicians in the United States and
many in Europe administer such treatment. I'm not set up to do this,
but I do ask my patients to take large oral doses of alpha Upoic acid
and EPO, as Usted in Chapter 15. Asindicatedin that chapter,I suggest
biotin supplementation whenever alpha lipoic acid is used. The prob
lem here is that at the doses listed on page 241 (1,800 mg ALA daily),
the users must take 9-12 daily piUs over and above whatever other
medications or supplements they may be taking. Nevertheless, I con
tinue to prescribe these supplements for those who can afford them in
the hope that vagal healingcanbe accelerated, but I don't reaUy expect
a miracle.
Asmentioned earUer in this book, many diabeticshave another en
docrine disorder, hypothyroidism. Since diminished production of
thyroid hormones can cause neuropathy even in nondiabetics, it
wouldbe appropriate for diabetics with neuropathyof the vagusnerve
(gastroparesis) to be tested for thyroid insufficiency. If this turns out
to be present,the treatment is usuaUy 1piU daily. Aneasy curefor gas
troparesis, if it was not causedbyhigh blood sugars.
380 Treatment
Don't hesitate to use combinations ofthe medications and other treat
ments forgastroparesis that we have covered in this chapter. The more
methods you findthat wiUwork for you, the betterthe likelyoutcome.
There is one exceptionto this rule— do notuse both domperidone and
metoclopramide. Use only one or the other, as they both work by the
same mechanism and their potential for adverse effects wiU increase
with the combined dosage.
The effects upon blood sugar of even asymptomatic (symptom-
free) delayed stomach-emptying from any cause can be dramatic.
Don't think that because you have no symptoms you're free from its
effects upon blood sugar. If you're uncertain, ask your physician to
perform an R-R interval study. If you're foUowing the guidelines of
this book and your blood sugars arestill unpredictable,suggestthat he
or she readthis chapter.
Routine Follow-up Visits
to Your Physician
Taking responsibiUty for thecare of yourown diabetes may free
you from habits that have beenwithyou for manyyears. It also
requires the estabUshment of new habits, such as exercise and
bloodsugar self-monitoring, that are easier to abandon thanto foUow.
Once yourbloodsugars have become controUed, it mayonlytakea
few months for youconveniendy to forget about the pain youusedto
have in your toes, or the parent or friend who lost alegor vision due
to compUcations of diabetes, andsoon. Astime goes on, youwiU find
thatwithdiabetes, as withlife ingeneral, youwiU graduaUy tendtodo
what is easiest or most enjoyable at the moment. This backsliding is
quite common. When I haven't seen a patient for six months, I'll
usuaUy take a meal history and find that some of the basic dietary
guidelines have been forgotten. Concurrendy blood sugar profiles,
glycosylated hemoglobin levels, Upid profiles, andeven fibrinogen lev
els may have deteriorated. Such deterioration can be short-circuited
when I see patients every two months. We aU need alittle nudge to get
back on track, and it seems that a time frame of about two months
does the trick for most of us. I was not the first diabetologist to ob
serve this, and your physician maylikewise want you to visit him at
simUar intervals.
Dosage requirements for insulin or ISAs may change over time,
whether due to weight changes, to deterioration or improvement of
betaceU output, or just to seasonal temperature changes. Sothere's an
ongoing need for readjustment of these medications. Again, two-
month intervals are appropriate.
What are someof the things that yourphysician maywant to con
sider at these foUow-up visits?
382 Treatment
First of aU, your doctorshouldtry to answer anynewquestionsthat
youmayhave. These maycover ahost of subjects, from somethingyou
readin the newspaper to new physical complaints or dissatisfaction
with your diet. Write down your questions in advance, so that you
won't forget them.
Your physician wiU, of course, want to reviewyourblood sugar data
sheets covering a period of at least two weeks. It makes no sense for
your doctor to review prior data, as that is old history. If he or she
wantsto adjustyour medications or meal plan, the changes shouldbe
based upon current information. Remember, however, that the data
must be complete and honest. This means, for example, that if you
spent afew hoursshopping oroverate, it shouldbenotedon yourdata
sheet. It doesn't make sense, and can be dangerous, for your doctor to
change your medications based upon high blood sugars caused by a
fewunrecorded dietaryindiscretions.
Yourphysician wiUalso wantto drawsomeblood.At each visit your
HgbAlc (glycated hemoglobin) should be checked. You need not be
fasting for this test. Up-to-date physicians are now performing this
test in the office using a smaU drop of finger-stick blood. Resultscan
be had in about 6 minutes. At least once annuaUy, a complete lipid
profile including LDL subparticles should be performed, and fibrino
genlevels should be checked; C-reactive protein also should be mea
sured. Kidney function studies including crystatin-c should also be
performed. You'U recaU that these require a 24-hoururine coUection,
which must be completedon the day of the visit (see Chapter 2). Re
member that the "normal values" for lipid profiles are based upon
fasting determinations. Soif yourphysician hasplanned suchtests, try
to book anearly-morning appointment, anddon't eatbreakfast. If you
skipbreakfast, be sure also to skip your preprandial insulinor ISAif
youusuaUy usethese medications to cover breakfast. Donot omit glu
cose tablets or Humalog (Uspro) needed to correct low or elevated
blood sugars. Alsoremember to take your basal dose of ISA or long-
acting insuUn, as their purpose is merely to hold blood sugar level
whUe fasting. Your physician may also want to perform other blood
tests from time to time, such as ablood count and a chemical profile.
If you are takingastatin drug for elevated levels of smaU dense LDL,
Uver function tests should be performed.
A partial physical examination, including weight, should be per
formed every two months. UsuaUy the most important element of
RoutineFollow-up Visitsto Your Physician 383
these visits should be examination of your feet. Such an examination
is not merely to look for injuries, bUsters, or what have you. EquaUy
important is the discovery of dry skin, athlete's foot, pressure points
from iU-fitting shoes, ingrown or fungus-infected toenaUs, and cal
luses. Your shoes should also be examined for areas where they have
been stretchedby prominences on your toes, suggesting that they are
smaUer than your feet. Any of these can cause or may indicate prob
lems that could leadto ulcersof the feet and should be corrected. Dry
skin is best treatedwith daUy appUcations of animal or vegetable oUs
such as vitamin EoU, olive oU, emulsified lanoUn, mink oU, emu oU, or
anyproprietary oU otherthan mineraloU. The cureforiU-fitting shoes
is new shoes(possiblycustom-made) with awide toe box with adeep
rise. CaUuses frequendy requirethe purchase of custom orthotics that
redistributethe pressure on the bottoms of your feet. Grindingoff cal
luses is not the solution, as caUuses are a symptom, not a cause, of
excess pressure. Their removal is the most common causeof amputa
tions in patients that I seeat my hospital'swound careclinic.
Resting blood pressures, repeated every few minutes until the low
est reading isobtained,are mandatoryat everyvisit if yourblood pres
sureis even sUghdy elevated. If yourblood pressure is usuaUy normal,
it should be checked everytwelvemonths anyway.
Over the courseof ayear or two, other aspects of physical examina
tion should be performed. The tests need not be done aU at one visit,
but may be staggered. These include osciUometric studies ofthe blood
circulation in your legs, an electrocardiogram, tests for sensation in
your feet, andacompleteeyeexam. The eyeexamshouldincludepupU-
laryreflexes, visual acuity, intraocular pressure, the Amslergridtest, a
test for doublevision, and examination of your lenses, anteriorcham
bers, and retinas through dUated pupUs. This last exam must be per
formed with certainspecialized equipment that should include direct
andindirect ophthalmoscopes andaslitlamp. If your physician is not
so equipped, or if he has previouslyfound potentialvision-threatening
changes in your eyes, you shouldbe referred to anophthalmologist or
If your initial physical examdisclosed diabetic complications such
as earlysigns of neuropathy, carpal tunnel syndrome, or Dupuytren's
contractures,examination for these compUcations should be periodi-
caUy repeated. The R-R interval study should be repeated every eigh
teen months, even if it was initiaUy normal.
384 Treatment
The best treatment for the computations of diabetes is prevention.
The second best treatment is detection in the very early stages, whue
reversal is still possible. For theseand the reasons mentioned above, I
stronglyrecommendvisits to your physician every two months, or at
least everythree months.
What You Can Expect from Virtually
Normal Blood Sugars
I amconvinced frommypersonal experience, from the experiences
of my patients, and from reading the scientific Uterature, that
people with normal bloodsugars do not develop the long-term
compUcations of diabetes. I am further convinced that diabetics with
even slightly elevated blood glucose profiles may eventuaUy experi
ence someof thelong-termconsequences of diabetes, but theywiU de
velop moreslowly andlikely beless severe thanforpeople withhigher
blood sugars. Inthischapter, I wiU trytodescribe some of thechanges
that I and other physicians have observed when the blood sugars of
our patients dramaticaUy improve.
Most common, perhaps, is the feeling of being more alert and no
longer chronicaUy tired. Many people who "feel perfectly fine" before
their blood sugars are normalized comment later that they had no
idea that they could feelso much better.
Another common occurrence relates to short-term memory. Very
frequentiy patients or spouses wiU refer to a patient's"terriblemem
ory." When I first began mymedical practice, I would askpatients to
phone meat nightwith theirblood sugar dataforfine-tuning of med
ications. My wife, aphysician specializing inpsychoanalytic medicine,
sometimes overheard my end of the conversation and would com
ment, "Thatperson hasadementia." Weeks later, shewould again hear
my end of a conversation with the same individual, and would com
ment on the great improvement of short-termmemory. This became
386 Treatment
so common that I introduced an objectivetest for short-term memory
into the neurologic exam that I performon aU newpatients.* More
thanhalfmynewpatients indeed display thismUd formof dementia,
which appears to Uft after several months of improved blood sugar.
The improvement is usuaUy quite apparent to spouses.
Diabetic neuropathies seemto improve in twophases — a rapid par
tial improvement that may occur within weeks, foUowed bysustained
veryslowimprovement that goes onforyears ifbloodsugars continue
to remain normal. Thisis most apparent with numbness or pain in the
toes. Some people wiU even comment, "I knowright away if myblood
sugaris high, because mytoes feel numb again." On the other hand,
several patientswith total numbness of their feet have complained of
severe pain after several months of near-normal blood sugars. This
continues for a number of months and eventuaUy resolves as sensa
tion returns. It is as if nerves generate pain signals whUe they heal or
"sprout." The experience may be very frightening and distressing if
you haven'tbeenwarnedthat it might occur.
Erectile dysfunction affects about 60percentof diabetic males, and
is the result of years of elevated blood sugars. It maybe definedas an
inabUity to maintaina rigid enough penUe erection for adequate time
to perform intercourse. It usuaUy results from neuropathy, blocked
blood vessels, or both. We can perform simple tests to determine
whichof thesecauses predominates. Whenthe problemis principaUy
neurologic, I frequently hearthecomment, sometimes afteronlya few
weeks of near-normal blood sugar profiles, "Hey, I'm able to have in
tercourse again!" Unfortunately, this turnaround only appears to oc
cur if the man was able to attain at least partial erections before. If at
the original interview, I'mtold, "Doc, it'sbeen dead foryears," I know
recovery isunlikely tooccur. Iftesting shows that theproblemwas due
primarily to blocked bloodvessels, I never see improvement. Note,
however, that it's normal to be unable to have erections when blood
sugars aretoo low, say below80mg/dl.
*I recite sixdigits (Sam Spade's license number) and askthe patient to repeat
them in reverse order.
What You Can Expect from Virtually Normal Blood Sugars 387
Another remarkable change relates to autonomic neuropathy and
associated gastroparesis. I have documented major improvement of
R-R interval studies in many patients, and total normalization in a
few. Along with this, we see reduction insigns and symptoms of gas
troparesis. UsuaUy such improvement takes place over a period of
years. Although it occurs most dramaticaUy in younger people, I've
also seen it occurin seventy-year-olds.
Diplopia, or double vision, iscaused byneuropathy of thenerves that
activate themuscles that move theeyes. It isaverycommon finding in
the physical examination that I perform, but rarely severe enough to
benoticed by patients onaday-to-day basis. Here again, when testing
is redone after a few years, we find improvement or even total cures
with blood sugar improvement.
Vacuoles are tiny bubbles in thelens of theeye andare thought to
be precursors of cataracts. I haveseen a number of these vanish after a
year or two of improved blood sugars. I have even seen thedisappear
ance of smaU"spokes" on thelens that signify very early cataracts.
I've seen mUd cases of glaucoma cured bynormalization of blood
sugars, as weU as retinal hemorrhages, macular edema, and micro
Improvements in risk factors forheartdisease, such asmildhyperten
sion, elevated cholesterol/HDL ratios, triglycerides, and fibrinogen
levels, are commonplace. They usuaUy can be observed after about
two months of sustained normal or near-normal blood sugars and
continue to improve for about oneyear.
Sinularly, improvements in early changes noted on renal risk pro
files are often obtained, usuaUy after one or two years, but sometimes
after a few months.
It has long been known that elevated blood sugars adversely affect
growth in chUdren and teenagers. As blood sugars approach nor
mal, chUdren with delayed growth rapidly return to theirprediabetic
growth curves. I, unfortunately, missed this opportunitybecause I was
388 Treatment
thirty-nine years oldwhen I finaUy figured out howto normalize my
blood sugars. I did have, however, the joyof watching my nondiabetic
son and some of my young diabetic patients become giants in com
parison to me.
Most dramatic and commonplace is the feeUng of satisfactionand
control that nearlyeveryone experiences when they produce normal
or nearly normal bloodsugar profiles. This is especiaUy true for indi
vidualswho hadalready beentakinginsulin,but appears also to occur
in those who do not take insulin.
In the late 1970s, the methods of this book were used at the Rocke-
feUer University to normalizeblood sugars in a group of type 1 dia
betics.They wereinitiaUy testedby a psychiatrist usingthe HamUton
depression scale. The starting score for the group was in the"severely
depressed" range. This dropped to normal after the patients became
the masters of their blood sugars.
Lastbut not least is the feeling that we are not doomed to share the
fate of others we haveknown, who died prematurelyafteryears of dis-
abUng or painful diabetic complications. We come to reaUze that with
the abiUty to controlour blood sugars comesthe abiUty to preventthe
consequences of high blood sugars.
I have long maintained that diabetics are entided to the same blood
sugars as nondiabetics. But it is up to us to see that we achieve this
Your Diabetic Cookbook
Sauces 394
Breakfast Foods 398
Soups 401
Salads 405
Poultry 407
Beef, Lamb, and Veal 411
Pork 416
Seafood 417
Vegetable Entrees
and Side Dishes 424
Quiches and Souffles 437
Desserts 439
Recipes for LowCarbohydrate Meals
The recipes that foUow are in and of themselves wonderful ex
amples of howyou can eat weUwithveryUttle fast-acting car
bohydrate.They are, however, not intended asthe end-aU and
be-aU for diabetic nutrition. As youlearned in Chapters 9-11, devel
oping a meal plan is at its foundation science, but there is also art
involved. The science offersyou the metaboUc and nutritional under
pinnings of what should andshould not bein yourmeal plan. The art
portion isthe negotiation thathas totake place between youandyour
physician, and between your nutritional needs and your Ufestyle, es-
peciaUy your tastes andthe time you have to spend in cooking. You
can do weU with these recipes, but you can also do weU by adjusting
these recipes to your own tastes. The recipes were developed by two
quite talented but different chefs. Karen A. Weinstock, who wrote
most of the newerrecipes in thisbook, isherselfatype 1diabetic. She
isalso anutritional health care provider. For more than twentyyears,
she has taught cooking to individuals, with the goal of maximizing
health through diet. When she was diagnosed with type 1 diabetes
many years ago, her sense of healthy eating went into atailspin, asher
own diet, like that of most Americans — even those who "know
food" — included an excess of refined and fast-acting carbohydrate.
After many years of unsuccessfiiUy regulating her own blood sugar
levels, she met me. She says, "This program saved my life by providing
me with the necessary guidelines andpractical how-tosto livea nor
mal life as a diabetic."
Her goal in creating recipes for this book was to providethe dia
betic communitywithdeUcious gourmet meals based on our program
andher own nutritional expertise. She says, "It is my hope that you
392 Your Diabetic Cookbook
can enjoy both preparing and eating these meals whUe maintaining
healthyblood sugar levels." The recipes she created are identified by
the initials kw at the end. The recipesidentifiedby the initials ta were
createdbyTimothyJ. Aubert, CWC, for the first editionof this book.
AU the recipes are, in one sense, a guide to howyou can incorporate
intoyour diet foods youmaynot have considered eating, and howyou
can use low-carbohydratefoods and protein to arrive at tasty approx
imations of foods fromthe high-carbohydrate world.
You can use the recipes exacdy as written and trust that they wiU
playa significant rolein assisting youwithbloodsugarnormalization;
or you canplaywiththemand customize them, to suit your own tastes
and dietary guidelines. It is best, however, unless you are a seasoned
cookyourself, to trytherecipes first astheyarewritten,andthen make
adjustments if they seem warranted. Changes in herbs and spices or
includingsUghtiymorewhole-plant vegetables that arelisted in the"So
What's Left to Eat?"section in Chapter 10are not likelyto alter blood
sugars significandy, but you should foUow carbohydrate and protein
content guidelines and check your blood sugars to make sure. If a
recipe calls for less carbohydrate than required byyour meal plan, add
somevegetables, salad, bran crackers, et cetera, to the meal to make up
the difference. Referto Chapter 10for some typical suggestions.
If you've flipped straight to these recipes without gaining a good
understanding of how to foUow a meal plan, stop and at least read
Chapters 9-11. Look especiaUy at theboxentided"No-No's in a Nut-
shett," on pages 152-153. Then look at the list of vegetables on page
151; it's likelythat any vegetable not Usted in that section is not suit
ablylowin fast-acting carbohydrate. Remember that Vi cup of diced
or sUced cookedlow-carbohydrate vegetable (or V4 cup mashed) is ap
proximately equivalent to 6 grams carbohydrate, as is 1cup of mixed
salad. Assume that %cup of whole cookedvegetables is also equiva
lent to about 6 grams of carbohydrate.
Throughout these recipes the abbreviation cho is used for carbo
hydrate (cho stands for carbon, hydrogen, and oxygen, the elements
that make up carbohydrates) and pro for protein. Each recipe shows
the number of servings provided and the approximate grams of car
bohydratesand ouncesof proteinin eachserving. (If youare adapting
Recipes for Low-Carbohydrate Meals 393
these recipes or creating your ownand consulting food value books,
remember our rule of thumb, that to convert grams of protein to
ounces of a raw protein food, you divide by6. Divide by9 for cooked
protein foods.)
As you know, thepaper packets containing granulated, so-caUed sugar-
free sweeteners usuaUy contain about 96 percent glucose, maltodex-
trin, or other sugar, making theminappropriate for diabetics. You can
prepare your own granulated sweetener for use in some of the foUow
ing recipes by crushing or grinding aspartame or saccharin tablets
(not packets) in oneof the foUowing ways:
• in a mortar and pesde
• betweentwo spoons
• in a pepper miU
• in a smaU electric coffee grinder
You can also dissolve the crushed tablets in a smaU amount of hot
water (unless the recipe caUs for powdered sweetener).
Aspartame (but not saccharin) wiUloseits taste if added to food be
fore cooking, soit must beused only after cooking. You may prefer to
use stevia, since it is sold as powder or liquid andis not degraded by
heat. Make surethat youonlypurchase powdered stevia that does not
contain maltodextrin.
A number of these recipes include bran crackers. Some have been
written for Wasa Fiber Rye crackers (5 grams carbohydrate each),
some for Bran-a-Crisp crackers (4 grams carbohydrate each), and
others for G/G Scandinavian Bran Crispbread (3 grams carbohydrate
each). Although these crackers are simUar intexture, they vary incar
bohydrate value. Make theproper food count adjustment ifyou should
swap Wasa, say, for G/G.
Just because you're diabetic doesn't mean you can'teat weU. In fact,
you are likely toeatbetter than theaverage person who husties abagel
394 Your Diabetic Cookbook
or donut for breakfast, scrounges fast foodfor lunch, then doeswhat
ever for dinner.
Sobon app&it, salut, and enjoy.
Mybook The Diabetes Diet, also pubUshed by Littie, Brown, has a
hundred newlow-carbohydrate recipes for breakfast, lunch, and din
ner designed specificaUy for bloodsugarnormalization.
6 servings, about Vi cup each Perserving: 5.6gmcho, <0.5 0Z PRO
CHO(gm) PRO (gm)
3 cups diced red beU peppers 27.6 3.6
1Tbsp oliveoU
— —
Va cup chopped freshbasU
0.4 0.2
2 cloves garlic, minced
2.0 0.4
1 cup CoUege Inn chicken broth
— 1.0
Vi cup heavycream
2.2 1.6
Vi tsp salt
— —
Vs tsp black pepper
— —
l/i tsp dried oregano
1.0 0.2
l/i tsp steviapowder (1 packet)
— —
2TbspgratedParmesan cheese 0.4 4.2
Inasaucepan, bring aquart ofwater toaboU. Add diced peppers, cover,
and simmer for 20 minutes. Drain Uquid from peppers by pouring
through acolander. Add peppers tofood processor workbowl andpuree
for 2-3 minutes. The finished texture of the puree wiU contain some
pulp. Heat oUve oU inasaucepan over alowflame. Add basUandgarUc.
Saute" on a lowflame until the aroma is released, 3-4 minutes. Stir in the
pepper puree, chicken broth, andheavy cream. WhUe stirring, addre
maining seasonings except forgrated cheese. Simmer thesauce uncov
eredfor 40minutes. Addgratedcheese to sauce just beforeserving.
This recipe yields about 3cupsof sauce.
Thissauce canbeused inmanyrecipes that caUfor a redsauce, such
as meatioaf or stuffedcabbage. It wiU keepin the refrigeratorfor 4-5
days. It may bestored in thefreezer for 2-3 months, kw
Recipesfor Low-CarbohydrateMeals 395
16servings, about 1 Tbsp each Perserving: 1.2 gm cho, <0.1 OZPRO
CHOfem) PRO(gm)
2 cups diced red beU pepper 18.4 2.4
lA tsp salt — —
Vi tsp steviapowder (1 packet) — —
2 tsp white horseradish
— —
1Tbsp Worcestershire sauce 1.0

1Tbsp cider vinegar — —
Hot sauce to taste
— —
In a saucepan, bring a quart of water to a boU. Add diced red pepper
and cover the pot. Reduceheat to a simmer and cook for 15 minutes.
Drain water from peppers and put them into a blender.Blendpeppers
to a pureed consistency. Pour puree into a glass bowl and refrigerate
for 30minutes. Remove the chiUed pepper puree fromthe refrigerator.
Add the salt and steviato the puree and combine. Awhiskis helpful
here. Add horseradish, Worcestershire sauce, vinegar, and hot sauce.
Serve this sauce with chiUed shrimp or other seafood. This recipe
yields 1cup of cocktaU sauce.
The saucewiU keep in the refrigerator for 2-3 daysin a glass jar. It
may be stored in the freezer for 2-3 months, kw
24servings, 1Tbsp each Per serving: <1gmcho, <0.1 oz pro
CHO (gm) PRO (gm)
1Vi cups diced red beU pepper 13.8 1.8
%cup mayonnaise — —
1Tbsp canola oU — —
1 tsp Worcestershire sauce 0.3 —
Vi tsp stevia powder (1 packet) — —
Va tsp black pepper — —
2 Tbsp dicedsour diU pickle 0.5 —
In a saucepan, bring 1quart ofwater to a boU. Add peppers and cover.
Reduce heat to simmerfor 10minutes. Drainpeppers in a colanderand
discardcookingUquid. Addpeppersto the workbowl of a food proces
sor.Chop for 2-3 minutes. The consistencyshouldbesmooth, but some
of the fiber wiU remain. Add the mayonnaise, oU, and Worcestershire
sauce. Blend together for a moment. Add the stevia and black pepper
and blend again. Scrape mixture from the workbowl into a glassbowl.
Fold in diced pickle. This dressing wiU keep in the refrigerator for
10-14 days. (Don't try to freeze.) Storein an airtight container, kw
396 Your Diabetic Cookbook
12servings, 2 Tbsp each Per serving: (17gm cho, <0.1 OZPRO
CHO(gm) PRO(gm)
2 Tbsp minced shaUots 3.4 0.6
\lAcups {2l/isticks) butter, softened —
3 Tbsp Dijon mustard 3.0 —
1tsp lemon juice 0.43 0.03
1Tbsp Worcestershire sauce 1.0

Tabasco sauce to taste 0.5

SauteshaUots in 1teaspoonof thebutter. In a foodprocessor, combine
shaUots with aU other ingredients until smooth. Place on parchment
paper or on plasticwrap. RoU butter in the paper or wrap until you
have a 1-inch-diameter cylinder. Refrigerate until butter is needed.
Slice into V4-inch pieces to use (3sUces = 1tablespoon), ta
4 servings, 2 Tbsp each Per serving: 0.7gmcho, <0.1 ozpro
CHO (gm) PRO(gm)
Vi cup (1 stick) unsalted butter, softened — 1.0
2 Tbsp lemon juice 2.6 0.2
Salt and white pepper to taste — —
In a food processor,combine aU ingredients until smooth. RoU butter
into a 1-inch-diameter cylinder as directed for Dijon Mustard Butter,
above, and refrigerate untU needed. SUce into V4-inch pieces to use
(3 sUces = 1tablespoon), ta
4 servings, 2 Tbsp each Per serving: 0.7gmcho, <0.1 ozpro
CHO (gm) PRO (gm)
Vi cup (1 stick) butter, softened — 1.0
2 Tbsp lemon juice 2.6 0.2
Vs tsp salt — —
Ys tsp white pepper — —
Lemon pepper seasoning, to taste — —
In a food processor, combine aU ingredients until smooth. RoU butter
into a 1-inch-diameter cylinder as directed for Dijon Mustard Butter,
above, and refrigerate until needed. SUce into V4-inch pieces to use
(3 sUces = 1 tablespoon), ta
Recipesfor Low-Carbohydrate Meals 397
12servings, 2 Tbspeach Perserving: 0.5 gm cho, <0.1 OZPRO
CHO(gm) PRO(gm)
Wa cups {2Vi sticks) butter, softened —
4 minced scaUions 1.85 0.45
Va tsp minced garUc 0.5 0.08
Vi tsp minced freshginger 0.15 0.01
1Tbsp minced parsley 0.6 0.3
1Tbsp soysauce 2.0 2.0
1Tbsp lemon juice 1.3 0.1
In a food processor, combineaU ingredients until smooth. RoU butter
into a 1-inch-diametercylinder as directedfor Dijon Mustard Butter,
above, and refrigerate until needed. Slice into V4-inch pieces to use (3
sUces = 1tablespoon), ta
12servings, 2 Tbsp each Per serving: 1.7gmcho, <0.5 oz pro
Ya cup white wine
1bay leaf
7 black peppercorns, crushed
3 Tbsp tarragon vinegar
2 Tbsp minced shaUots
1 cup heavy cream
Va cup (IVz sticks) butter, softened
V4 tsp chopped fresh tarragon
Pinch salt and white pepper — —
Combine wine, bayleaf, crushedpeppercorns, vinegar, and shaUots in
a nonreactive pan. Bringto a boU and reduceto about 2 tablespoons.
Strain, removing bayleafand crushed peppercorns. Reduce creamby
halfin a separate panandaddto thewine reduction. GraduaUy whisk
in the butter over low heat. When aUthe butter is dissolved, add the
chopped freshtarragon and seasonto taste, ta
16servings, 2 Tbsp each Per serving: 2.3gmcho, <0.5oz pro
CHO (gm) PRO(gm)
V2 cup minced shaUots 19.2 2.8
Va cup cider vinegar — —
1tsp black peppercorns, cracked 1.4 0.2
IO(gm) PRO(gm)
7.2 0.6
0.3 0.1
0.7 0.1
4.8 0.7
6.6 4.9

0.8 0.4
398 Your Diabetic Cookbook
1bay leaf 0.3 0.1
1 cup dry white wine 9.6 —
1 cup heavycream 6.6 4.9
Wi cups (3 sticks) unsaltedbutter, softened — 3.0
Dijon-style mustard to taste — —
Creolemustard (or anyother spicymustard)
to taste
— —
Combine the first 5 ingredients in a smaU nonreactive saucepan and
reduce to Vi cup. Add heavy creamand reduce mixture by half. Strain
and return to the stove. Cut softenedbutter into smaUpiecesand slowly
add to the sauce whUe whisking. After aU the butter is incorporated,
add the mustard to taste, ta
Breakfast is the meal where you mayfindyou miss carbohy
drate the most. No more home fries or hash browns, toast,
pancakes, French toast, waffles, cereals, and the like. The
recipe suggestions that follow can put some zip back into
breakfast while keeping carbohydrates waydown.
1serving Perserving: 3.1gm cho, 2.8 oz pro
CHO (gm) PRO (gm)
2 sUces bacon — 4.0
1 fresh mushroom, sUced 1.5 0.35
Butter to taste — —
2 eggs 1.2 12.0
1Tbsp heavycream 0.4 0.3
Salt and black pepper to taste — —
Pan-frybacon and removeto paper towelto drain. Saute" sUced mush
room in butter for 2-3 minutes. In a smaUbowl, mix eggswith cream,
then add to mushrooms. Cookeggs without stirring for 2 minutes, or
until desired firmness. Season with salt and pepper to taste. RoU or
fold omelet and turn out on a plate. Servewith the bacon, ta
Recipesfor Low-Carbohydrate Meals 399
1serving Perserving5.3gmcho, 2.4 ozpro
CHO (gm) PRO(gm)
2 sUces Stripples (soybacon) 2.0 2.0
2 eggs 1.2 12.0
1Tbsp cream 0.4 0.3
Butter to taste — —
1Tbsp minced onion 0.9 0.3
1Tbsp minced greenbeU pepper 0.8 —
Salt and blackpepper to taste — —
Microwave Stripples and set aside. Combine eggs and cream thor
oughlyin a smatt bowl. Heat butter in saute" pan, add eggs, and cook
for 1 minute. Add minced onion and green pepper. Seasonwith salt
and pepper to taste and cook to desiredconsistency, ta
1 serving Perserving. 3•6gm CHO, 6 oz PRO
CHO(gm) PRO(gm)
2 eggs 1.2 12.0
1Tbsp cream 0.4 0.3
Butter to taste
— —
1sUce (2 oz) ham, diced or juUenned —
2 oz cheese, gratedor sUced thin 2.0 12.0
Salt and black pepper to taste — —
Mixtogether eggsand creamin a smaU bowl. Heat butter in saute" pan
and cook egg mixture 1-2 minutes without stirring. Place ham and
cheese on top and seasonto taste with salt and pepper. Either roU or
foldthe eggs and cookto desired consistency, ta
This recipewasdeveloped byAmyZ. Kornfeld and Hank Kornfeld.
J serving Perserving: 8gmcho, 5 ozpro
CHO (gm) PRO (gm)
2 sausage patties, 1 oz each, or
2 slices ham, turkey, or salami — 12.0
1Tbspbutter or 1tsp vegetable oU — —
2 eggs 1.2 12.0
400 Your Diabetic Cookbook
2 G/G crispbreads 6.0 2.0
2 slices cheese (about Va ounce total) 0.8 4.0
Brownsausage,ham, turkey,or salami and drain off fat. Keep warm in
250°F oven. Heat butter or oU in a nonstick skiUet until water drops
sprinkled on surface skitter across. Breakeggs into pan. Fry eggs for
2-3 minutes over medium heat. If desired, flip them over and fry for
another minute or so. Put crispbreads on an ovenproofplateand place
eggs on top. Cover with sausage, ham, turkey, or salami, and top off
with cheese. Warmbrieflyin ovento melt cheese, ta
This is another recipefrom AmyZ. Kornfeldand Hank Kornfeld.
1serving Perserving: 9 gmcho, 1.4oz pro
CHO (gm) PRO(gm)
2 Bran-a-Crispcrackers 8.0 2.0
2 tsp water — —
1 eggor eggsubstitute 0.6 6.0
Va tsp cinnamon — —
V% tsp nutmeg — —
Vs tsp vaniUa extract — —
Artificial maple or fruit-flavored baking
extract, to taste — —
1Tbsp cream 0.4 0.3
1 tsp vegetable oU — —
Melted butter to taste — —
1 or more ground or crushed Equal tablets,
or a pinch of stevia powder — —
Soakcrackersin 2 teaspoonswater for 5 minutes, or just long enough
to soften. MeanwhUe, in a broad shaUow bowl beat eggor eggsubsti
tute with cinnamon, nutmeg, vaniUa, and other flavor extract. Add
1Tbsp creamand beat gendy. Place softenedwafers in egg mixture for
1-2 minutes. Heat nonstick skiUet until water droplets sprinkled on
surface skitter across. Add oUto skiUet and spread it around with a
folded paper towel. Place egg-soaked wafers in pan and cook over
medium heat for about 3 minutes per side. When done remove from
pan and pour on meltedbutter, or sprinkle to taste with the ground
Equal tablets or steviapowder.
Recipesfor Low-Carbohydrate Meals 401
AmyZ. Kornfeldand Hank Kornfeld alsosuggested
this substitute for traditional pancakes.
1serving Perserving: 7 gmcho, 1.4ozpro
CHO (gm) PRO(gm)
2 G/G crispbreads 6.0 2.0
1egg,beaten 0.6 6.0
Vs tsp nutmeg — —
Va tsp cinnamon — —
Vz tsp vaniUa extract — —
Artificial vaniUa, orange, or almondbaking
extract, to taste — —
1Tbsp cream 0.4 0.3
1tsp vegetable oU — —
Melted butter to taste — —
1or more Equal tablets, ground, or pinch
stevia powder — —
Grindcrispbreads inblender, food processor, or electric coffee grinder
to a flourUke consistency. Combine egg, nutmeg, cinnamon, vaniUa,
baking extract, and cream in bowl. Add groundcrispbreads and mix.
Heat nonstickskiUet. Whenhot, addoil to skiUet and spreadit around
with a paper towel. Add one-quarter of batter to skiUet. Cook for
2 minutes. Turn carefuUy and cook other side for another 2 minutes,
to producefirst pancake. Repeat 3 timesto produce3 more pancakes.
Cover with melted butter. Sprinkle pancakes with ground Equal or
stevia powder.
4 servings, about 1cupeach Per serving: 8gmcho, <0.5 oz pro
CHO (gm) PRO(gm)
1smaU (1 lb) acorn squash (seeNote) 21.4 1.6
1Tbsp butter — —
2 stalkscelery, thinly sUced diagonaUy
(set aside leaves for garnish) 3.0 0.6
1smattleek, cleanedand thinlysUced
diagonaUy 4.0 0.4
Vi tsp salt — —
Vi cup heavycream 3.3 2.5
402 Your Diabetic Cookbook
V/a cups water — —
Va tsp cinnamon 0.4 —
Quarter the acorn squashand scrapeout the seeds. Usesome caution
when cutting the squash —the skin can be tough and the flesh is
dense, so work your knifein graduaUy as you cut.
In a largesaucepan,bring 5 cups of water to a boU. Addsquash and
simmer until tender, about 15minutes. Drainliquid fromsquashand
aUow to cool. Scoop pulp from the outer peel into the workbowl of
your food processor. Discardpeel. Puree cooked squash in the food
processor untU liquefied. This should yield about 1 cup of squash
puree (anymore wiU increase the carbohydrate count).
In the same saucepan, add butter and saute* celeryand leeks with
salt for 5-7 minutes, or until wUted. Addsquashpuree, heavy cream,
and water and stir weU. Heat on a lowflame for 10 minutes. Stir in cin
namon, garnish with the fresh celery leaves, and serve warm. The
bisque wiU keepin the refrigeratorfor 2-3 days.It maybe stored in the
freezer fori month.
This is a recipethat, because of its texture, can be fairlyeasfly di
gestedby those who suffer from gastroparesis.
Acorn squash has a wonderful flavor, verycloseto sweetpotatoes, with lots of
carotenoids but with considerablylessfast-actingcarbohydrate. Youcan sub
stitute canned pumpkin for the squash, but be sure to read food labels and
ensure that what you buy doesn't haveadded sugar (it doesn't as long as the
only ingredient is pumpkin), kw
4 servings, about 3A cupeach Perserving: 3.9gm cho, <0.5 oz pro
CHO (gm) PRO(gm)
1 whole cucumber, peeled and sUced
Vi cup chopped fennel
1Tbsp whole-milk yogurt
Vi cup cold water — —
Vi tsp fresh diU — —
Lemon pepper seasoning to taste — —
In a blender combine sliced cucumber, fennel, yogurt, water, and
diU. Puree untU smooth, season with lemon pepper seasoning, and
serve, ta
8.3 2.1
6.3 1.1
0.75 0.75
Recipes for Low-Carbohydrate Meals 403
3 servings, about 1¥i cups each Per serving: 3.2gmcho, <0.5oz pro
CHO (gm) PRO(gm)
4 medium zucchini, cleaned and sliced 4.0 3.5
1lhTbsp chopped onion 1.35 0.15
1Tbsp butter — 0.1
3 Tbsp hot water — —
1 cube Knorr's chicken bouiUon, crushed 2.0 0.8
Salt, blackpepper, and garUc powderto taste — —
xh cup heavycream 2.2 1.63
In a 2-quart pan, saute zucchini and onion in butter until tender.
Transfer vegetables to a blenderand puree.Addhot waterwith crushed
bouiUon cube; blend for 1-2 minutes. Season with salt, pepper, and
garUc powder to taste. Serve hot or cold. Add cream to individual por
tions, about 2 tablespoonsper serving, ta
4 servings Per serving: 3.5gmcho, 2.3 oz pro
CHO(gm) PRO (gm)
2 large eggs 1.2 12.4
4 cups CoUege Inn chicken broth —
Vi cup finely chopped scaUions 3.7 0.9
6 oz cooked chicken breast fiUet, shredded
(see Note 1) —
1Tbsp soysauce 2.0 2.0
1Tbsp arrowroot powder (seeNote 2) 7.0

2 Tbsp cold water — —
In a mixing bowl, crackopen eggs and whisk togetherwith Vi cup of
the chicken broth. Place this mixture in the refrigerator. Place the re
maining 3Vi cups chicken broth in a saucepan. Bring liquid to a low
boU. Add scaUions, shredded chicken, andsoysauce. Reduce heat and
simmer 3-5 minutes. Remove egg mixture from the refrigerator.
Slowly addegg mixtureto thesoup, while stirringthe soupconstantly.
As you stir in the egg, it wiU feather out and thicken. Dissolve arrow
root powder in coldwaterina smaUbowl. Stirthisintothe soup.Cook
for 3-5 minutes witha lid. Serve warm. This recipe wiU keep in the re
frigerator for 4-5 days. It may also be stored in the freezer for 2-3
404 Your Diabetic Cookbook
1. Leftovercooked chickenworks best for this recipe. It is easiest to shred
chickenwhen it is cold, usinga fork or tongs.
2. Arrowroot powder (alsocalled arrowroot flour) is a thickening agent
used like cornstarch. It must be dissolved in a small amount of cold water be
fore being added to warm ingredients, kw
6 servings Perserving: 2.9gm CHO, 2.8 OZ PRO
CHO (gm) PRO (gm)
Vi lb shrimp in their sheUs 0 31.6
3 cups water
— —
2 Tbsp butter — —
Vi cup thinly slicedbutton mushrooms 4.0 1.7
1 stalk celery, diced 1.5 0.3
Vi tsp dried tarragon 0.4 —
— —
1tsp salt — —
Vi tsp black pepper
— —
Vi lb scrod, cod, or other firm white fish — 40.4
Vi cup heavycream 3.3 2.4
Vi cup dry white wine
4.8 —
Vi lb bay scaUops 3.6 25.3
To clean shrimp, rinse in a colander under cold running water. Re
move sheUs and set them aside. They wiU be used for the soup stock.
Remove the veinalongthe back and discardit. Cut shrimp into smaU
bite-sizedpieces and set aside. Place sheUs in a saucepanwith 3 cups of
water. Bringto a boUandsimmerfor 10minutes. Strainout sheUs and
return stock to saucepan. In a skiUet, heat butter on a lowflame. Add
mushrooms, celery, tarragon, bayleaf, salt, andblackpepper. Saute" for
3-5 minutes. MeanwhUe, add scrod, heavycream, and white wine to
the seafood stock. Add the cooked vegetables from the skiUet. Turn on
heat under saucepan and bring chowderto a lowboU. The scrod wiU
faU apart asit cooks, sothereisno needto cut it. Reduce heat and sim
mer covered for 15 minutes. Add scaUops and shrimp to chowder.
Cook for 3-5 minutes. Remove bay leaf, adjust seasoning, and serve
The chowder wiU keepin the refrigeratorfor 3-4 days. It may also
be stored in the freezer for 1 month, kw
Recipesfor Low-Carbohydrate Meals 405
4 servings Perserving: 3 gm CHO, 4 OZPRO
CHO (gm) PRO(gm)
1 lb chicken breasts, boned and skinned —
1stalk celery, diced 1.5 0.3
Vi cup coarselygrated jicama (seeNote) 5.3 0.4
Vi cup diced green beU pepper 4.6 0.6
3 Tbsp cider vinegar — —
3 Tbsp mayonnaise — —
Vi tsp salt — —
Vi tsp blackpepper — —
Va tsp stevia powder (Vi packet) — —
Vi tsp curry powder 0.6 0.1
To cookchicken breasts, steamthem until tender and no longer pink
inside, 12-15minutes.AUowto coolbefore sUcing. Cut chicken breasts
into bite-sized chunks.
In a medium-sizedglass mixingbowl, combinechicken and celery,
jicama,and greenpepper.Addvinegarand mayonnaise and mixthor
oughly. Add salt, pepper, stevia, and curry power to chicken salad.
Blendin aU seasoning by evenly coatingthe ingredients. Serve chiUed
or at room temperature.
If you're not famttiar with jicama, it has brown skin like a potato and is
shaped something likea turnip. Its flesh hasa pleasing crunch, almostlikea
crispapple. When purchasing jicama, which is available thesedays in most
supermarkets, makesurethe skinis firmwithout bruises. The smaU variety,
about the sizeof an orange, is the best for its taste and texture, kw
4 servings Perserving: 9.4gmcho, 0.5oz pro
CHO (gm) PRO(gm)
1eggplant,about Wa lb, peeledand cut
into lV^-inch cubes 27.8 4.67
%tsp salt — —
5 Tbsp olive oU — —
Va tsp dried oregano 0.2 —
2 clovesgarUc, minced 2.0 0.2
Va tsp black pepper — —
1 oz {Va cup) crumbled feta cheese 1.0 5.0
406 YourDiabetic Cookbook
2 Tbsp lemon juice 2.6 0.2
1 small head of Boston lettuce, washed,
dried, andUghtiy shredded 3.8 2.1
Preheat oven to 425°F. Rub eggplant chunks with salt and aUow to
stand for 10minutes. With apapertowel, pat dry anysurface moisture
(this helpsto removethe bittertaste). In alarge mixingbowl, combine
the eggplant, 4 Tbsp olive oU (set aside 1Tbsp olive oU for final sea
soning), oregano, garUc, andblack pepper. Mixthoroughly and place
in a9 x 13 bakingdish. Bake uncovered for 45minutes. Remove from
oven and aUow to cool. Sprinklewith feta cheese, lemon juice, and 1
Tbsp oUve oU beforeserving on abed of Bostonlettuce.
Serve chiUedor at room temperature, kw
4 servings Per serving. 2.6gmcho, <0.5oz pro
CHO (gm) PRO(gm)
4 picklingcucumbers, sliced into
y3-inch rounds 5.6 1.6
2 stalkscelery, sUced in t